walkero
/
pcre2
1
0
Fork 0
Code Issues Pull Requests Projects Releases 1 Wiki Activity

Partial match documentation rewritten.

This commit is contained in:
Philip.Hazel 2019-08-07 17:21:02 +00:00
parent 59c7c5d100
commit ce751bfc84
3 changed files with 752 additions and 924 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

575
doc/html/pcre2partial.html
View File

@ -14,85 +14,123 @@ please consult the man page, in case the conversion went wrong.
<br>
<ul>
<li><a name="TOC1" href="#SEC1">PARTIAL MATCHING IN PCRE2</a>
<li><a name="TOC2" href="#SEC2">PARTIAL MATCHING USING pcre2_match()</a>
<li><a name="TOC3" href="#SEC3">PARTIAL MATCHING USING pcre2_dfa_match()</a>
<li><a name="TOC4" href="#SEC4">PARTIAL MATCHING AND WORD BOUNDARIES</a>
<li><a name="TOC5" href="#SEC5">EXAMPLE OF PARTIAL MATCHING USING PCRE2TEST</a>
<li><a name="TOC2" href="#SEC2">REQUIREMENTS FOR A PARTIAL MATCH</a>
<li><a name="TOC3" href="#SEC3">PARTIAL MATCHING USING pcre2_match()</a>
<li><a name="TOC4" href="#SEC4">MULTI-SEGMENT MATCHING WITH pcre2_match()</a>
<li><a name="TOC5" href="#SEC5">PARTIAL MATCHING USING pcre2_dfa_match()</a>
<li><a name="TOC6" href="#SEC6">MULTI-SEGMENT MATCHING WITH pcre2_dfa_match()</a>
<li><a name="TOC7" href="#SEC7">MULTI-SEGMENT MATCHING WITH pcre2_match()</a>
<li><a name="TOC8" href="#SEC8">ISSUES WITH MULTI-SEGMENT MATCHING</a>
<li><a name="TOC9" href="#SEC9">AUTHOR</a>
<li><a name="TOC10" href="#SEC10">REVISION</a>
<li><a name="TOC7" href="#SEC7">AUTHOR</a>
<li><a name="TOC8" href="#SEC8">REVISION</a>
</ul>
<br><a name="SEC1" href="#TOC1">PARTIAL MATCHING IN PCRE2</a><br>
<P>
In normal use of PCRE2, if the subject string that is passed to a matching
function matches as far as it goes, but is too short to match the entire
pattern, PCRE2_ERROR_NOMATCH is returned. There are circumstances where it
might be helpful to distinguish this case from other cases in which there is no
match.
In normal use of PCRE2, if there is a match up to the end of a subject string,
but more characters are needed to match the entire pattern, PCRE2_ERROR_NOMATCH
is returned, just like any other failing match. There are circumstances where
it might be helpful to distinguish this "partial match" case.
</P>
<P>
Consider, for example, an application where a human is required to type in data
for a field with specific formatting requirements. An example might be a date
in the form <i>ddmmmyy</i>, defined by this pattern:
One example is an application where the subject string is very long, and not
all available at once. The requirement here is to be able to do the matching
segment by segment, but special action is needed when a matched substring spans
the boundary between two segments.
</P>
<P>
Another example is checking a user input string as it is typed, to ensure that
it conforms to a required format. Invalid characters can be immediately
diagnosed and rejected, giving instant feedback.
</P>
<P>
Partial matching is a PCRE2-specific feature; it is not Perl-compatible. It is
requested by setting one of the PCRE2_PARTIAL_HARD or PCRE2_PARTIAL_SOFT
options when calling a matching function. The difference between the two
options is whether or not a partial match is preferred to an alternative
complete match, though the details differ between the two types of matching
function. If both options are set, PCRE2_PARTIAL_HARD takes precedence.
</P>
<P>
If you want to use partial matching with just-in-time optimized code, as well
as setting a partial match option for the matching function, you must also call
<b>pcre2_jit_compile()</b> with one or both of these options:
<pre>
^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
</pre>
If the application sees the user's keystrokes one by one, and can check that
what has been typed so far is potentially valid, it is able to raise an error
as soon as a mistake is made, by beeping and not reflecting the character that
has been typed, for example. This immediate feedback is likely to be a better
user interface than a check that is delayed until the entire string has been
entered. Partial matching can also be useful when the subject string is very
long and is not all available at once, as discussed below.
</P>
<P>
PCRE2 supports partial matching by means of the PCRE2_PARTIAL_SOFT and
PCRE2_PARTIAL_HARD options, which can be set when calling a matching function.
The difference between the two options is whether or not a partial match is
preferred to an alternative complete match, though the details differ between
the two types of matching function. If both options are set, PCRE2_PARTIAL_HARD
takes precedence.
</P>
<P>
If you want to use partial matching with just-in-time optimized code, you must
call <b>pcre2_jit_compile()</b> with one or both of these options:
<pre>
PCRE2_JIT_PARTIAL_SOFT
PCRE2_JIT_PARTIAL_HARD
PCRE2_JIT_PARTIAL_SOFT
</pre>
PCRE2_JIT_COMPLETE should also be set if you are going to run non-partial
matches on the same pattern. If the appropriate JIT mode has not been compiled,
interpretive matching code is used.
matches on the same pattern. Separate code is compiled for each mode. If the
appropriate JIT mode has not been compiled, interpretive matching code is used.
</P>
<P>
Setting a partial matching option disables two of PCRE2's standard
optimizations. PCRE2 remembers the last literal code unit in a pattern, and
abandons matching immediately if it is not present in the subject string. This
optimization cannot be used for a subject string that might match only
partially. PCRE2 also knows the minimum length of a matching string, and does
optimization hints. PCRE2 remembers the last literal code unit in a pattern,
and abandons matching immediately if it is not present in the subject string.
This optimization cannot be used for a subject string that might match only
partially. PCRE2 also remembers a minimum length of a matching string, and does
not bother to run the matching function on shorter strings. This optimization
is also disabled for partial matching.
</P>
<br><a name="SEC2" href="#TOC1">PARTIAL MATCHING USING pcre2_match()</a><br>
<br><a name="SEC2" href="#TOC1">REQUIREMENTS FOR A PARTIAL MATCH</a><br>
<P>
A partial match occurs during a call to <b>pcre2_match()</b> when the end of the
subject string is reached successfully, but matching cannot continue because
more characters are needed, and in addition, either at least one character in
the subject has been inspected or the pattern contains a lookbehind, or (when
PCRE2_PARTIAL_HARD is set) the pattern could match an empty string. An
inspected character need not form part of the final matched string; lookbehind
assertions and the \K escape sequence provide ways of inspecting characters
before the start of a matched string.
A possible partial match occurs during matching when the end of the subject
string is reached successfully, but either more characters are needed to
complete the match, or the addition of more characters might change what is
matched.
</P>
<P>
The three additional requirements define the cases where adding more characters
to the existing subject may complete the same match that would occur if they
had all been present in the first place. Without these conditions there would
be a partial match of an empty string at the end of the subject for all
unanchored patterns (and also for anchored patterns if the subject itself is
empty).
Example 1: if the pattern is /abc/ and the subject is "ab", more characters are
definitely needed to complete a match. In this case both hard and soft matching
options yield a partial match.
</P>
<P>
Example 2: if the pattern is /ab+/ and the subject is "ab", a complete match
can be found, but the addition of more characters might change what is
matched. In this case, only PCRE2_PARTIAL_HARD returns a partial match;
PCRE2_PARTIAL_SOFT returns the complete match.
</P>
<P>
On reaching the end of the subject, when PCRE2_PARTIAL_HARD is set, if the next
pattern item is \z, \Z, \b, \B, or $ there is always a partial match.
Otherwise, for both options, the next pattern item must be one that inspects a
character, and at least one of the following must be true:
</P>
<P>
(1) At least one character has already been inspected. An inspected character
need not form part of the final matched string; lookbehind assertions and the
\K escape sequence provide ways of inspecting characters before the start of a
matched string.
</P>
<P>
(2) The pattern contains one or more lookbehind assertions. This condition
exists in case there is a lookbehind that inspects characters before the start
of the match.
</P>
<P>
(3) There is a special case when the whole pattern can match an empty string.
When the starting point is at the end of the subject, the empty string match is
a possibility, and if PCRE2_PARTIAL_SOFT is set and neither of the above
conditions is true, it is returned. However, because adding more characters
might result in a non-empty match, PCRE2_PARTIAL_HARD returns a partial match,
which in this case means "there is going to be a match at this point, but until
some more characters are added, we do not know if it will be an empty string or
something longer".
</P>
<br><a name="SEC3" href="#TOC1">PARTIAL MATCHING USING pcre2_match()</a><br>
<P>
When a partial matching option is set, the result of calling
<b>pcre2_match()</b> can be one of the following:
</P>
<P>
<b>A successful match</b>
A complete match has been found, starting and ending within this subject.
</P>
<P>
<b>PCRE2_ERROR_NOMATCH</b>
No match can start anywhere in this subject.
</P>
<P>
<b>PCRE2_ERROR_PARTIAL</b>
Adding more characters may result in a complete match that uses one or more
characters from the end of this subject.
</P>
<P>
When a partial match is returned, the first two elements in the ovector point
@ -110,26 +148,6 @@ these characters are needed for a subsequent re-match with additional
characters.
</P>
<P>
What happens when a partial match is identified depends on which of the two
partial matching options is set.
</P>
<br><b>
PCRE2_PARTIAL_SOFT WITH pcre2_match()
</b><br>
<P>
If PCRE2_PARTIAL_SOFT is set when <b>pcre2_match()</b> identifies a partial
match, the partial match is remembered, but matching continues as normal, and
other alternatives in the pattern are tried. If no complete match can be found,
PCRE2_ERROR_PARTIAL is returned instead of PCRE2_ERROR_NOMATCH.
</P>
<P>
This option is "soft" because it prefers a complete match over a partial match.
All the various matching items in a pattern behave as if the subject string is
potentially complete. For example, \z, \Z, and $ match at the end of the
subject, as normal, and for \b and \B the end of the subject is treated as a
non-alphanumeric.
</P>
<P>
If there is more than one partial match, the first one that was found provides
the data that is returned. Consider this pattern:
<pre>
@ -138,26 +156,34 @@ the data that is returned. Consider this pattern:
If this is matched against the subject string "abc123dog", both alternatives
fail to match, but the end of the subject is reached during matching, so
PCRE2_ERROR_PARTIAL is returned. The offsets are set to 3 and 9, identifying
"123dog" as the first partial match that was found. (In this example, there are
two partial matches, because "dog" on its own partially matches the second
alternative.)
"123dog" as the first partial match. (In this example, there are two partial
matches, because "dog" on its own partially matches the second alternative.)
</P>
<br><b>
PCRE2_PARTIAL_HARD WITH pcre2_match()
How a partial match is processed by pcre2_match()
</b><br>
<P>
If PCRE2_PARTIAL_HARD is set for <b>pcre2_match()</b>, PCRE2_ERROR_PARTIAL is
returned as soon as a partial match is found, without continuing to search for
possible complete matches. This option is "hard" because it prefers an earlier
partial match over a later complete match. For this reason, the assumption is
made that the end of the supplied subject string may not be the true end of the
available data, and so, if \z, \Z, \b, \B, or $ are encountered at the end
of the subject, the result is PCRE2_ERROR_PARTIAL, whether or not any
characters have been inspected.
What happens when a partial match is identified depends on which of the two
partial matching options is set.
</P>
<P>
If PCRE2_PARTIAL_HARD is set, PCRE2_ERROR_PARTIAL is returned as soon as a
partial match is found, without continuing to search for possible complete
matches. This option is "hard" because it prefers an earlier partial match over
a later complete match. For this reason, the assumption is made that the end of
the supplied subject string is not the true end of the available data, which is
why \z, \Z, \b, \B, and $ always give a partial match.
</P>
<P>
If PCRE2_PARTIAL_SOFT is set, the partial match is remembered, but matching
continues as normal, and other alternatives in the pattern are tried. If no
complete match can be found, PCRE2_ERROR_PARTIAL is returned instead of
PCRE2_ERROR_NOMATCH. This option is "soft" because it prefers a complete match
over a partial match. All the various matching items in a pattern behave as if
the subject string is potentially complete; \z, \Z, and $ match at the end of
the subject, as normal, and for \b and \B the end of the subject is treated
as a non-alphanumeric.
</P>
<br><b>
Comparing hard and soft partial matching
</b><br>
<P>
The difference between the two partial matching options can be illustrated by a
pattern such as:
@ -182,26 +208,132 @@ to follow this explanation by thinking of the two patterns like this:
The second pattern will never match "dogsbody", because it will always find the
shorter match first.
</P>
<br><a name="SEC3" href="#TOC1">PARTIAL MATCHING USING pcre2_dfa_match()</a><br>
<br><b>
Example of partial matching using pcre2test
</b><br>
<P>
The DFA functions move along the subject string character by character, without
The <b>pcre2test</b> data modifiers <b>partial_hard</b> (or <b>ph</b>) and
<b>partial_soft</b> (or <b>ps</b>) set PCRE2_PARTIAL_HARD and PCRE2_PARTIAL_SOFT,
respectively, when calling <b>pcre2_match()</b>. Here is a run of
<b>pcre2test</b> using a pattern that matches the whole subject in the form of a
date:
<pre>
re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data&#62; 25dec3\=ph
Partial match: 23dec3
data&#62; 3ju\=ph
Partial match: 3ju
data&#62; 3juj\=ph
No match
</pre>
This example gives the same results for both hard and soft partial matching
options. Here is an example where there is a difference:
<pre>
re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data&#62; 25jun04\=ps
0: 25jun04
1: jun
data&#62; 25jun04\=ph
Partial match: 25jun04
</pre>
With PCRE2_PARTIAL_SOFT, the subject is matched completely. For
PCRE2_PARTIAL_HARD, however, the subject is assumed not to be complete, so
there is only a partial match.
</P>
<br><a name="SEC4" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre2_match()</a><br>
<P>
PCRE was not originally designed with multi-segment matching in mind. However,
over time, features (including partial matching) that make multi-segment
matching possible have been added. The string is searched segment by segment by
calling <b>pcre2_match()</b> repeatedly, with the aim of achieving the same
results that would happen if the entire string was available for searching.
</P>
<P>
Special logic must be implemented to handle a matched substring that spans a
segment boundary. PCRE2_PARTIAL_HARD should be used, because it returns a
partial match at the end of a segment whenever there is the possibility of
changing the match by adding more characters. The PCRE2_NOTBOL option should
also be set for all but the first segment.
</P>
<P>
When a partial match occurs, the next segment must be added to the current
subject and the match re-run, using the <i>startoffset</i> argument of
<b>pcre2_match()</b> to begin at the point where the partial match started.
Multi-segment matching is usually used to search for substrings in the middle
of very long sequences, so the patterns are normally not anchored. For example:
<pre>
re&#62; /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
data&#62; ...the date is 23ja\=ph
Partial match: 23ja
data&#62; ...the date is 23jan19 and on that day...\=offset=15
0: 23jan19
1: jan
</pre>
Note the use of the <b>offset</b> modifier to start the new match where the
partial match was found.
</P>
<P>
In this simple example, the next segment was just added to the one in which the
partial match was found. However, if there are memory constraints, it may be
necessary to discard text that precedes the partial match before adding the
next segment. In cases such as the above, where the pattern does not contain
any lookbehinds, it is sufficient to retain only the partially matched
substring. However, if a pattern contains a lookbehind assertion, characters
that precede the start of the partial match may have been inspected during the
matching process.
</P>
<P>
The only lookbehind information that is available is the length of the longest
lookbehind in a pattern. This may not, of course, be at the start of the
pattern, but retaining that many characters before the partial match is
sufficient, if not always strictly necessary. The way to do this is as follows:
</P>
<P>
Before doing any matching, find the length of the longest lookbehind in the
pattern by calling <b>pcre2_pattern_info()</b> with the PCRE2_INFO_MAXLOOKBEHIND
option. Note that the resulting count is in characters, not code units. After a
partial match, moving back from the ovector[0] offset in the subject by the
number of characters given for the maximum lookbehind gets you to the earliest
character that must be retained. In a non-UTF or a 32-bit situation, moving
back is just a subtraction, but in UTF-8 or UTF-16 you have to count characters
while moving back through the code units. Characters before the point you have
now reached can be discarded.
</P>
<P>
For example, if the pattern "(?&#60;=123)abc" is partially matched against the
string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maximum
lookbehind count is 3, so all characters before offset 2 can be discarded. The
value of <b>startoffset</b> for the next match should be 3. When <b>pcre2test</b>
displays a partial match, it indicates the lookbehind characters with '&#60;'
characters if the <b>allusedtext</b> modifier is set:
<pre>
re&#62; "(?&#60;=123)abc"
data&#62; xx123ab\=ph,allusedtext
Partial match: 123ab
&#60;&#60;&#60;
</pre>
Note that the \fPallusedtext\fP modifier is not available for JIT matching,
because JIT matching does not maintain the first and last consulted characters.
</P>
<br><a name="SEC5" href="#TOC1">PARTIAL MATCHING USING pcre2_dfa_match()</a><br>
<P>
The DFA function moves along the subject string character by character, without
backtracking, searching for all possible matches simultaneously. If the end of
the subject is reached before the end of the pattern, there is the possibility
of a partial match, again provided that at least one character has been
inspected.
of a partial match.
</P>
<P>
When PCRE2_PARTIAL_SOFT is set, PCRE2_ERROR_PARTIAL is returned only if there
have been no complete matches. Otherwise, the complete matches are returned.
However, if PCRE2_PARTIAL_HARD is set, a partial match takes precedence over
any complete matches. The portion of the string that was matched when the
longest partial match was found is set as the first matching string.
If PCRE2_PARTIAL_HARD is set, a partial match takes precedence over any
complete matches. The portion of the string that was matched when the longest
partial match was found is set as the first matching string.
</P>
<P>
Because the DFA functions always search for all possible matches, and there is
no difference between greedy and ungreedy repetition, their behaviour is
different from the standard functions when PCRE2_PARTIAL_HARD is set. Consider
the string "dog" matched against the ungreedy pattern shown above:
Because the DFA function always searches for all possible matches, and there is
no difference between greedy and ungreedy repetition, its behaviour is
different from the <b>pcre2_match()</b>. Consider the string "dog" matched
against this ungreedy pattern:
<pre>
/dog(sbody)??/
</pre>
@ -209,58 +341,16 @@ Whereas the standard function stops as soon as it finds the complete match for
"dog", the DFA function also finds the partial match for "dogsbody", and so
returns that when PCRE2_PARTIAL_HARD is set.
</P>
<br><a name="SEC4" href="#TOC1">PARTIAL MATCHING AND WORD BOUNDARIES</a><br>
<P>
If a pattern ends with one of sequences \b or \B, which test for word
boundaries, partial matching with PCRE2_PARTIAL_SOFT can give counter-intuitive
results. Consider this pattern:
<pre>
/\bcat\b/
</pre>
This matches "cat", provided there is a word boundary at either end. If the
subject string is "the cat", the comparison of the final "t" with a following
character cannot take place, so a partial match is found. However, normal
matching carries on, and \b matches at the end of the subject when the last
character is a letter, so a complete match is found. The result, therefore, is
<i>not</i> PCRE2_ERROR_PARTIAL. Using PCRE2_PARTIAL_HARD in this case does yield
PCRE2_ERROR_PARTIAL, because then the partial match takes precedence.
</P>
<br><a name="SEC5" href="#TOC1">EXAMPLE OF PARTIAL MATCHING USING PCRE2TEST</a><br>
<P>
If the <b>partial_soft</b> (or <b>ps</b>) modifier is present on a
<b>pcre2test</b> data line, the PCRE2_PARTIAL_SOFT option is used for the match.
Here is a run of <b>pcre2test</b> that uses the date example quoted above:
<pre>
re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data&#62; 25jun04\=ps
0: 25jun04
1: jun
data&#62; 25dec3\=ps
Partial match: 23dec3
data&#62; 3ju\=ps
Partial match: 3ju
data&#62; 3juj\=ps
No match
data&#62; j\=ps
No match
</pre>
The first data string is matched completely, so <b>pcre2test</b> shows the
matched substrings. The remaining four strings do not match the complete
pattern, but the first two are partial matches. Similar output is obtained
if DFA matching is used.
</P>
<P>
If the <b>partial_hard</b> (or <b>ph</b>) modifier is present on a
<b>pcre2test</b> data line, the PCRE2_PARTIAL_HARD option is set for the match.
</P>
<br><a name="SEC6" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre2_dfa_match()</a><br>
<P>
When a partial match has been found using a DFA matching function, it is
When a partial match has been found using the DFA matching function, it is
possible to continue the match by providing additional subject data and calling
the function again with the same compiled regular expression, this time setting
the PCRE2_DFA_RESTART option. You must pass the same working space as before,
because this is where details of the previous partial match are stored. Here is
an example using <b>pcre2test</b>:
because this is where details of the previous partial match are stored. You can
set the PCRE2_PARTIAL_SOFT or PCRE2_PARTIAL_HARD options with PCRE2_DFA_RESTART
to continue partial matching over multiple segments. Here is an example using
<b>pcre2test</b>:
<pre>
re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data&#62; 23ja\=dfa,ps
@ -272,143 +362,10 @@ The first call has "23ja" as the subject, and requests partial matching; the
second call has "n05" as the subject for the continued (restarted) match.
Notice that when the match is complete, only the last part is shown; PCRE2 does
not retain the previously partially-matched string. It is up to the calling
program to do that if it needs to.
</P>
<P>
That means that, for an unanchored pattern, if a continued match fails, it is
not possible to try again at a new starting point. All this facility is capable
of doing is continuing with the previous match attempt. In the previous
example, if the second set of data is "ug23" the result is no match, even
though there would be a match for "aug23" if the entire string were given at
once. Depending on the application, this may or may not be what you want.
The only way to allow for starting again at the next character is to retain the
matched part of the subject and try a new complete match.
</P>
<P>
You can set the PCRE2_PARTIAL_SOFT or PCRE2_PARTIAL_HARD options with
PCRE2_DFA_RESTART to continue partial matching over multiple segments. This
facility can be used to pass very long subject strings to the DFA matching
functions.
</P>
<br><a name="SEC7" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre2_match()</a><br>
<P>
Unlike the DFA function, it is not possible to restart the previous match with
a new segment of data when using <b>pcre2_match()</b>. Instead, new data must be
added to the previous subject string, and the entire match re-run, starting
from the point where the partial match occurred. Earlier data can be discarded.
</P>
<P>
It is best to use PCRE2_PARTIAL_HARD in this situation, because it does not
treat the end of a segment as the end of the subject when matching \z, \Z,
\b, \B, and $. Consider an unanchored pattern that matches dates:
<pre>
re&#62; /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
data&#62; The date is 23ja\=ph
Partial match: 23ja
</pre>
At this stage, an application could discard the text preceding "23ja", add on
text from the next segment, and call the matching function again. Unlike the
DFA matching function, the entire matching string must always be available,
and the complete matching process occurs for each call, so more memory and more
processing time is needed.
</P>
<br><a name="SEC8" href="#TOC1">ISSUES WITH MULTI-SEGMENT MATCHING</a><br>
<P>
Certain types of pattern may give problems with multi-segment matching,
whichever matching function is used.
</P>
<P>
1. If the pattern contains a test for the beginning of a line, you need to pass
the PCRE2_NOTBOL option when the subject string for any call does start at the
beginning of a line. There is also a PCRE2_NOTEOL option, but in practice when
doing multi-segment matching you should be using PCRE2_PARTIAL_HARD, which
includes the effect of PCRE2_NOTEOL.
</P>
<P>
2. If a pattern contains a lookbehind assertion, characters that precede the
start of the partial match may have been inspected during the matching process.
When using <b>pcre2_match()</b>, sufficient characters must be retained for the
next match attempt. You can ensure that enough characters are retained by doing
the following:
</P>
<P>
Before doing any matching, find the length of the longest lookbehind in the
pattern by calling <b>pcre2_pattern_info()</b> with the PCRE2_INFO_MAXLOOKBEHIND
option. Note that the resulting count is in characters, not code units. After a
partial match, moving back from the ovector[0] offset in the subject by the
number of characters given for the maximum lookbehind gets you to the earliest
character that must be retained. In a non-UTF or a 32-bit situation, moving
back is just a subtraction, but in UTF-8 or UTF-16 you have to count characters
while moving back through the code units.
</P>
<P>
Characters before the point you have now reached can be discarded, and after
the next segment has been added to what is retained, you should run the next
match with the <b>startoffset</b> argument set so that the match begins at the
same point as before.
</P>
<P>
For example, if the pattern "(?&#60;=123)abc" is partially matched against the
string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maximum
lookbehind count is 3, so all characters before offset 2 can be discarded. The
value of <b>startoffset</b> for the next match should be 3. When <b>pcre2test</b>
displays a partial match, it indicates the lookbehind characters with '&#60;'
characters if the "allusedtext" modifier is set:
<pre>
re&#62; "(?&#60;=123)abc"
data&#62; xx123ab\=ph,allusedtext
Partial match: 123ab
&#60;&#60;&#60;
</pre>
However, the "allusedtext" modifier is not available for JIT matching, because
JIT matching does not maintain the first and last consulted characters.
</P>
<P>
3. Matching a subject string that is split into multiple segments may not
always produce exactly the same result as matching over one single long string
when PCRE2_PARTIAL_SOFT is used. The section "Partial Matching and Word
Boundaries" above describes an issue that arises if the pattern ends with \b
or \B. Another kind of difference may occur when there are multiple matching
possibilities, because (for PCRE2_PARTIAL_SOFT) a partial match result is given
only when there are no completed matches. This means that as soon as the
shortest match has been found, continuation to a new subject segment is no
longer possible. Consider this <b>pcre2test</b> example:
<pre>
re&#62; /dog(sbody)?/
data&#62; dogsb\=ps
0: dog
data&#62; do\=ps,dfa
Partial match: do
data&#62; gsb\=ps,dfa,dfa_restart
0: g
data&#62; dogsbody\=dfa
0: dogsbody
1: dog
</pre>
The first data line passes the string "dogsb" to a standard matching function,
setting the PCRE2_PARTIAL_SOFT option. Although the string is a partial match
for "dogsbody", the result is not PCRE2_ERROR_PARTIAL, because the shorter
string "dog" is a complete match. Similarly, when the subject is presented to
a DFA matching function in several parts ("do" and "gsb" being the first two)
the match stops when "dog" has been found, and it is not possible to continue.
On the other hand, if "dogsbody" is presented as a single string, a DFA
matching function finds both matches.
</P>
<P>
Because of these problems, it is best to use PCRE2_PARTIAL_HARD when matching
multi-segment data. The example above then behaves differently:
<pre>
re&#62; /dog(sbody)?/
data&#62; dogsb\=ph
Partial match: dogsb
data&#62; do\=ps,dfa
Partial match: do
data&#62; gsb\=ph,dfa,dfa_restart
Partial match: gsb
</pre>
4. Patterns that contain alternatives at the top level which do not all start
with the same pattern item may not work as expected when PCRE2_DFA_RESTART is
used. For example, consider this pattern:
program to do that if it needs to. This means that, for an unanchored pattern,
if a continued match fails, it is not possible to try again at a new starting
point. All this facility is capable of doing is continuing with the previous
match attempt. For example, consider this pattern:
<pre>
1234|3789
</pre>
@ -417,30 +374,18 @@ alternative is found at offset 3. There is no partial match for the second
alternative, because such a match does not start at the same point in the
subject string. Attempting to continue with the string "7890" does not yield a
match because only those alternatives that match at one point in the subject
are remembered. The problem arises because the start of the second alternative
matches within the first alternative. There is no problem with anchored
patterns or patterns such as:
<pre>
1234|ABCD
</pre>
where no string can be a partial match for both alternatives. This is not a
problem if a standard matching function is used, because the entire match has
to be rerun each time:
<pre>
re&#62; /1234|3789/
data&#62; ABC123\=ph
Partial match: 123
data&#62; 1237890
0: 3789
</pre>
Of course, instead of using PCRE2_DFA_RESTART, the same technique of re-running
the entire match can also be used with the DFA matching function. Another
possibility is to work with two buffers. If a partial match at offset <i>n</i>
in the first buffer is followed by "no match" when PCRE2_DFA_RESTART is used on
the second buffer, you can then try a new match starting at offset <i>n+1</i> in
the first buffer.
are remembered. Depending on the application, this may or may not be what you
want.
</P>
<br><a name="SEC9" href="#TOC1">AUTHOR</a><br>
<P>
If you do want to allow for starting again at the next character, one way of
doing it is to retain the matched part of the segment and try a new complete
match, as described for <b>pcre2_match()</b> above. Another possibility is to
work with two buffers. If a partial match at offset <i>n</i> in the first buffer
is followed by "no match" when PCRE2_DFA_RESTART is used on the second buffer,
you can then try a new match starting at offset <i>n+1</i> in the first buffer.
</P>
<br><a name="SEC7" href="#TOC1">AUTHOR</a><br>
<P>
Philip Hazel
<br>
@ -449,9 +394,9 @@ University Computing Service
Cambridge, England.
<br>
</P>
<br><a name="SEC10" href="#TOC1">REVISION</a><br>
<br><a name="SEC8" href="#TOC1">REVISION</a><br>
<P>
Last updated: 22 July 2019
Last updated: 07 August 2019
<br>
Copyright &copy; 1997-2019 University of Cambridge.
<br>

538
doc/pcre2.txt
View File

@ -5650,72 +5650,109 @@ NAME
PARTIAL MATCHING IN PCRE2
In normal use of PCRE2, if the subject string that is passed to a
matching function matches as far as it goes, but is too short to match
the entire pattern, PCRE2_ERROR_NOMATCH is returned. There are circum-
stances where it might be helpful to distinguish this case from other
cases in which there is no match.
In normal use of PCRE2, if there is a match up to the end of a subject
string, but more characters are needed to match the entire pattern,
PCRE2_ERROR_NOMATCH is returned, just like any other failing match.
There are circumstances where it might be helpful to distinguish this
"partial match" case.
Consider, for example, an application where a human is required to type
in data for a field with specific formatting requirements. An example
might be a date in the form ddmmmyy, defined by this pattern:
One example is an application where the subject string is very long,
and not all available at once. The requirement here is to be able to do
the matching segment by segment, but special action is needed when a
matched substring spans the boundary between two segments.
^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
Another example is checking a user input string as it is typed, to en-
sure that it conforms to a required format. Invalid characters can be
immediately diagnosed and rejected, giving instant feedback.
If the application sees the user's keystrokes one by one, and can check
that what has been typed so far is potentially valid, it is able to
raise an error as soon as a mistake is made, by beeping and not re-
flecting the character that has been typed, for example. This immediate
feedback is likely to be a better user interface than a check that is
delayed until the entire string has been entered. Partial matching can
also be useful when the subject string is very long and is not all
available at once, as discussed below.
PCRE2 supports partial matching by means of the PCRE2_PARTIAL_SOFT and
PCRE2_PARTIAL_HARD options, which can be set when calling a matching
function. The difference between the two options is whether or not a
partial match is preferred to an alternative complete match, though the
details differ between the two types of matching function. If both op-
tions are set, PCRE2_PARTIAL_HARD takes precedence.
Partial matching is a PCRE2-specific feature; it is not Perl-compati-
ble. It is requested by setting one of the PCRE2_PARTIAL_HARD or
PCRE2_PARTIAL_SOFT options when calling a matching function. The dif-
ference between the two options is whether or not a partial match is
preferred to an alternative complete match, though the details differ
between the two types of matching function. If both options are set,
PCRE2_PARTIAL_HARD takes precedence.
If you want to use partial matching with just-in-time optimized code,
you must call pcre2_jit_compile() with one or both of these options:
as well as setting a partial match option for the matching function,
you must also call pcre2_jit_compile() with one or both of these op-
tions:
PCRE2_JIT_PARTIAL_SOFT
PCRE2_JIT_PARTIAL_HARD
PCRE2_JIT_PARTIAL_SOFT
PCRE2_JIT_COMPLETE should also be set if you are going to run non-par-
tial matches on the same pattern. If the appropriate JIT mode has not
been compiled, interpretive matching code is used.
tial matches on the same pattern. Separate code is compiled for each
mode. If the appropriate JIT mode has not been compiled, interpretive
matching code is used.
Setting a partial matching option disables two of PCRE2's standard op-
timizations. PCRE2 remembers the last literal code unit in a pattern,
and abandons matching immediately if it is not present in the subject
string. This optimization cannot be used for a subject string that
might match only partially. PCRE2 also knows the minimum length of a
matching string, and does not bother to run the matching function on
shorter strings. This optimization is also disabled for partial match-
ing.
timization hints. PCRE2 remembers the last literal code unit in a pat-
tern, and abandons matching immediately if it is not present in the
subject string. This optimization cannot be used for a subject string
that might match only partially. PCRE2 also remembers a minimum length
of a matching string, and does not bother to run the matching function
on shorter strings. This optimization is also disabled for partial
matching.
REQUIREMENTS FOR A PARTIAL MATCH
A possible partial match occurs during matching when the end of the
subject string is reached successfully, but either more characters are
needed to complete the match, or the addition of more characters might
change what is matched.
Example 1: if the pattern is /abc/ and the subject is "ab", more char-
acters are definitely needed to complete a match. In this case both
hard and soft matching options yield a partial match.
Example 2: if the pattern is /ab+/ and the subject is "ab", a complete
match can be found, but the addition of more characters might change
what is matched. In this case, only PCRE2_PARTIAL_HARD returns a par-
tial match; PCRE2_PARTIAL_SOFT returns the complete match.
On reaching the end of the subject, when PCRE2_PARTIAL_HARD is set, if
the next pattern item is \z, \Z, \b, \B, or $ there is always a partial
match. Otherwise, for both options, the next pattern item must be one
that inspects a character, and at least one of the following must be
true:
(1) At least one character has already been inspected. An inspected
character need not form part of the final matched string; lookbehind
assertions and the \K escape sequence provide ways of inspecting char-
acters before the start of a matched string.
(2) The pattern contains one or more lookbehind assertions. This condi-
tion exists in case there is a lookbehind that inspects characters be-
fore the start of the match.
(3) There is a special case when the whole pattern can match an empty
string. When the starting point is at the end of the subject, the
empty string match is a possibility, and if PCRE2_PARTIAL_SOFT is set
and neither of the above conditions is true, it is returned. However,
because adding more characters might result in a non-empty match,
PCRE2_PARTIAL_HARD returns a partial match, which in this case means
"there is going to be a match at this point, but until some more char-
acters are added, we do not know if it will be an empty string or some-
thing longer".
PARTIAL MATCHING USING pcre2_match()
A partial match occurs during a call to pcre2_match() when the end of
the subject string is reached successfully, but matching cannot con-
tinue because more characters are needed, and in addition, either at
least one character in the subject has been inspected or the pattern
contains a lookbehind, or (when PCRE2_PARTIAL_HARD is set) the pattern
could match an empty string. An inspected character need not form part
of the final matched string; lookbehind assertions and the \K escape
sequence provide ways of inspecting characters before the start of a
matched string.
When a partial matching option is set, the result of calling
pcre2_match() can be one of the following:
The three additional requirements define the cases where adding more
characters to the existing subject may complete the same match that
would occur if they had all been present in the first place. Without
these conditions there would be a partial match of an empty string at
the end of the subject for all unanchored patterns (and also for an-
chored patterns if the subject itself is empty).
A successful match
A complete match has been found, starting and ending within this sub-
ject.
PCRE2_ERROR_NOMATCH
No match can start anywhere in this subject.
PCRE2_ERROR_PARTIAL
Adding more characters may result in a complete match that uses one
or more characters from the end of this subject.
When a partial match is returned, the first two elements in the ovector
point to the portion of the subject that was matched, but the values in
@ -5731,23 +5768,6 @@ PARTIAL MATCHING USING pcre2_match()
string "abc12", because all these characters are needed for a subse-
quent re-match with additional characters.
What happens when a partial match is identified depends on which of the
two partial matching options is set.
PCRE2_PARTIAL_SOFT WITH pcre2_match()
If PCRE2_PARTIAL_SOFT is set when pcre2_match() identifies a partial
match, the partial match is remembered, but matching continues as nor-
mal, and other alternatives in the pattern are tried. If no complete
match can be found, PCRE2_ERROR_PARTIAL is returned instead of
PCRE2_ERROR_NOMATCH.
This option is "soft" because it prefers a complete match over a par-
tial match. All the various matching items in a pattern behave as if
the subject string is potentially complete. For example, \z, \Z, and $
match at the end of the subject, as normal, and for \b and \B the end
of the subject is treated as a non-alphanumeric.
If there is more than one partial match, the first one that was found
provides the data that is returned. Consider this pattern:
@ -5756,23 +5776,31 @@ PARTIAL MATCHING USING pcre2_match()
If this is matched against the subject string "abc123dog", both alter-
natives fail to match, but the end of the subject is reached during
matching, so PCRE2_ERROR_PARTIAL is returned. The offsets are set to 3
and 9, identifying "123dog" as the first partial match that was found.
(In this example, there are two partial matches, because "dog" on its
own partially matches the second alternative.)
and 9, identifying "123dog" as the first partial match. (In this exam-
ple, there are two partial matches, because "dog" on its own partially
matches the second alternative.)
PCRE2_PARTIAL_HARD WITH pcre2_match()
How a partial match is processed by pcre2_match()
If PCRE2_PARTIAL_HARD is set for pcre2_match(), PCRE2_ERROR_PARTIAL is
returned as soon as a partial match is found, without continuing to
search for possible complete matches. This option is "hard" because it
prefers an earlier partial match over a later complete match. For this
reason, the assumption is made that the end of the supplied subject
string may not be the true end of the available data, and so, if \z,
\Z, \b, \B, or $ are encountered at the end of the subject, the result
is PCRE2_ERROR_PARTIAL, whether or not any characters have been in-
spected.
What happens when a partial match is identified depends on which of the
two partial matching options is set.
Comparing hard and soft partial matching
If PCRE2_PARTIAL_HARD is set, PCRE2_ERROR_PARTIAL is returned as soon
as a partial match is found, without continuing to search for possible
complete matches. This option is "hard" because it prefers an earlier
partial match over a later complete match. For this reason, the assump-
tion is made that the end of the supplied subject string is not the
true end of the available data, which is why \z, \Z, \b, \B, and $ al-
ways give a partial match.
If PCRE2_PARTIAL_SOFT is set, the partial match is remembered, but
matching continues as normal, and other alternatives in the pattern are
tried. If no complete match can be found, PCRE2_ERROR_PARTIAL is re-
turned instead of PCRE2_ERROR_NOMATCH. This option is "soft" because it
prefers a complete match over a partial match. All the various matching
items in a pattern behave as if the subject string is potentially com-
plete; \z, \Z, and $ match at the end of the subject, as normal, and
for \b and \B the end of the subject is treated as a non-alphanumeric.
The difference between the two partial matching options can be illus-
trated by a pattern such as:
@ -5799,27 +5827,129 @@ PARTIAL MATCHING USING pcre2_match()
The second pattern will never match "dogsbody", because it will always
find the shorter match first.
Example of partial matching using pcre2test
The pcre2test data modifiers partial_hard (or ph) and partial_soft (or
ps) set PCRE2_PARTIAL_HARD and PCRE2_PARTIAL_SOFT, respectively, when
calling pcre2_match(). Here is a run of pcre2test using a pattern that
matches the whole subject in the form of a date:
re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data> 25dec3\=ph
Partial match: 23dec3
data> 3ju\=ph
Partial match: 3ju
data> 3juj\=ph
No match
This example gives the same results for both hard and soft partial
matching options. Here is an example where there is a difference:
re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data> 25jun04\=ps
0: 25jun04
1: jun
data> 25jun04\=ph
Partial match: 25jun04
With PCRE2_PARTIAL_SOFT, the subject is matched completely. For
PCRE2_PARTIAL_HARD, however, the subject is assumed not to be complete,
so there is only a partial match.
MULTI-SEGMENT MATCHING WITH pcre2_match()
PCRE was not originally designed with multi-segment matching in mind.
However, over time, features (including partial matching) that make
multi-segment matching possible have been added. The string is searched
segment by segment by calling pcre2_match() repeatedly, with the aim of
achieving the same results that would happen if the entire string was
available for searching.
Special logic must be implemented to handle a matched substring that
spans a segment boundary. PCRE2_PARTIAL_HARD should be used, because it
returns a partial match at the end of a segment whenever there is the
possibility of changing the match by adding more characters. The
PCRE2_NOTBOL option should also be set for all but the first segment.
When a partial match occurs, the next segment must be added to the cur-
rent subject and the match re-run, using the startoffset argument of
pcre2_match() to begin at the point where the partial match started.
Multi-segment matching is usually used to search for substrings in the
middle of very long sequences, so the patterns are normally not an-
chored. For example:
re> /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
data> ...the date is 23ja\=ph
Partial match: 23ja
data> ...the date is 23jan19 and on that day...\=offset=15
0: 23jan19
1: jan
Note the use of the offset modifier to start the new match where the
partial match was found.
In this simple example, the next segment was just added to the one in
which the partial match was found. However, if there are memory con-
straints, it may be necessary to discard text that precedes the partial
match before adding the next segment. In cases such as the above, where
the pattern does not contain any lookbehinds, it is sufficient to re-
tain only the partially matched substring. However, if a pattern con-
tains a lookbehind assertion, characters that precede the start of the
partial match may have been inspected during the matching process.
The only lookbehind information that is available is the length of the
longest lookbehind in a pattern. This may not, of course, be at the
start of the pattern, but retaining that many characters before the
partial match is sufficient, if not always strictly necessary. The way
to do this is as follows:
Before doing any matching, find the length of the longest lookbehind in
the pattern by calling pcre2_pattern_info() with the
PCRE2_INFO_MAXLOOKBEHIND option. Note that the resulting count is in
characters, not code units. After a partial match, moving back from the
ovector[0] offset in the subject by the number of characters given for
the maximum lookbehind gets you to the earliest character that must be
retained. In a non-UTF or a 32-bit situation, moving back is just a
subtraction, but in UTF-8 or UTF-16 you have to count characters while
moving back through the code units. Characters before the point you
have now reached can be discarded.
For example, if the pattern "(?<=123)abc" is partially matched against
the string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maxi-
mum lookbehind count is 3, so all characters before offset 2 can be
discarded. The value of startoffset for the next match should be 3.
When pcre2test displays a partial match, it indicates the lookbehind
characters with '<' characters if the allusedtext modifier is set:
re> "(?<=123)abc"
data> xx123ab\=ph,allusedtext
Partial match: 123ab
<<<
Note that the allusedtext modifier is not available for JIT matching,
because JIT matching does not maintain the first and last consulted
characters.
PARTIAL MATCHING USING pcre2_dfa_match()
The DFA functions move along the subject string character by character,
The DFA function moves along the subject string character by character,
without backtracking, searching for all possible matches simultane-
ously. If the end of the subject is reached before the end of the pat-
tern, there is the possibility of a partial match, again provided that
at least one character has been inspected.
tern, there is the possibility of a partial match.
When PCRE2_PARTIAL_SOFT is set, PCRE2_ERROR_PARTIAL is returned only if
there have been no complete matches. Otherwise, the complete matches
are returned. However, if PCRE2_PARTIAL_HARD is set, a partial match
takes precedence over any complete matches. The portion of the string
that was matched when the longest partial match was found is set as the
are returned. If PCRE2_PARTIAL_HARD is set, a partial match takes
precedence over any complete matches. The portion of the string that
was matched when the longest partial match was found is set as the
first matching string.
Because the DFA functions always search for all possible matches, and
there is no difference between greedy and ungreedy repetition, their
behaviour is different from the standard functions when PCRE2_PAR-
TIAL_HARD is set. Consider the string "dog" matched against the un-
greedy pattern shown above:
Because the DFA function always searches for all possible matches, and
there is no difference between greedy and ungreedy repetition, its be-
haviour is different from the pcre2_match(). Consider the string "dog"
matched against this ungreedy pattern:
/dog(sbody)??/
@ -5828,60 +5958,16 @@ PARTIAL MATCHING USING pcre2_dfa_match()
"dogsbody", and so returns that when PCRE2_PARTIAL_HARD is set.
PARTIAL MATCHING AND WORD BOUNDARIES
If a pattern ends with one of sequences \b or \B, which test for word
boundaries, partial matching with PCRE2_PARTIAL_SOFT can give counter-
intuitive results. Consider this pattern:
/\bcat\b/
This matches "cat", provided there is a word boundary at either end. If
the subject string is "the cat", the comparison of the final "t" with a
following character cannot take place, so a partial match is found.
However, normal matching carries on, and \b matches at the end of the
subject when the last character is a letter, so a complete match is
found. The result, therefore, is not PCRE2_ERROR_PARTIAL. Using
PCRE2_PARTIAL_HARD in this case does yield PCRE2_ERROR_PARTIAL, because
then the partial match takes precedence.
EXAMPLE OF PARTIAL MATCHING USING PCRE2TEST
If the partial_soft (or ps) modifier is present on a pcre2test data
line, the PCRE2_PARTIAL_SOFT option is used for the match. Here is a
run of pcre2test that uses the date example quoted above:
re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data> 25jun04\=ps
0: 25jun04
1: jun
data> 25dec3\=ps
Partial match: 23dec3
data> 3ju\=ps
Partial match: 3ju
data> 3juj\=ps
No match
data> j\=ps
No match
The first data string is matched completely, so pcre2test shows the
matched substrings. The remaining four strings do not match the com-
plete pattern, but the first two are partial matches. Similar output is
obtained if DFA matching is used.
If the partial_hard (or ph) modifier is present on a pcre2test data
line, the PCRE2_PARTIAL_HARD option is set for the match.
MULTI-SEGMENT MATCHING WITH pcre2_dfa_match()
When a partial match has been found using a DFA matching function, it
When a partial match has been found using the DFA matching function, it
is possible to continue the match by providing additional subject data
and calling the function again with the same compiled regular expres-
sion, this time setting the PCRE2_DFA_RESTART option. You must pass the
same working space as before, because this is where details of the pre-
vious partial match are stored. Here is an example using pcre2test:
vious partial match are stored. You can set the PCRE2_PARTIAL_SOFT or
PCRE2_PARTIAL_HARD options with PCRE2_DFA_RESTART to continue partial
matching over multiple segments. Here is an example using pcre2test:
re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
data> 23ja\=dfa,ps
@ -5894,141 +5980,10 @@ MULTI-SEGMENT MATCHING WITH pcre2_dfa_match()
(restarted) match. Notice that when the match is complete, only the
last part is shown; PCRE2 does not retain the previously partially-
matched string. It is up to the calling program to do that if it needs
to.
That means that, for an unanchored pattern, if a continued match fails,
it is not possible to try again at a new starting point. All this fa-
cility is capable of doing is continuing with the previous match at-
tempt. In the previous example, if the second set of data is "ug23" the
result is no match, even though there would be a match for "aug23" if
the entire string were given at once. Depending on the application,
this may or may not be what you want. The only way to allow for start-
ing again at the next character is to retain the matched part of the
subject and try a new complete match.
You can set the PCRE2_PARTIAL_SOFT or PCRE2_PARTIAL_HARD options with
PCRE2_DFA_RESTART to continue partial matching over multiple segments.
This facility can be used to pass very long subject strings to the DFA
matching functions.
MULTI-SEGMENT MATCHING WITH pcre2_match()
Unlike the DFA function, it is not possible to restart the previous
match with a new segment of data when using pcre2_match(). Instead, new
data must be added to the previous subject string, and the entire match
re-run, starting from the point where the partial match occurred. Ear-
lier data can be discarded.
It is best to use PCRE2_PARTIAL_HARD in this situation, because it does
not treat the end of a segment as the end of the subject when matching
\z, \Z, \b, \B, and $. Consider an unanchored pattern that matches
dates:
re> /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
data> The date is 23ja\=ph
Partial match: 23ja
At this stage, an application could discard the text preceding "23ja",
add on text from the next segment, and call the matching function
again. Unlike the DFA matching function, the entire matching string
must always be available, and the complete matching process occurs for
each call, so more memory and more processing time is needed.
ISSUES WITH MULTI-SEGMENT MATCHING
Certain types of pattern may give problems with multi-segment matching,
whichever matching function is used.
1. If the pattern contains a test for the beginning of a line, you need
to pass the PCRE2_NOTBOL option when the subject string for any call
does start at the beginning of a line. There is also a PCRE2_NOTEOL op-
tion, but in practice when doing multi-segment matching you should be
using PCRE2_PARTIAL_HARD, which includes the effect of PCRE2_NOTEOL.
2. If a pattern contains a lookbehind assertion, characters that pre-
cede the start of the partial match may have been inspected during the
matching process. When using pcre2_match(), sufficient characters must
be retained for the next match attempt. You can ensure that enough
characters are retained by doing the following:
Before doing any matching, find the length of the longest lookbehind in
the pattern by calling pcre2_pattern_info() with the
PCRE2_INFO_MAXLOOKBEHIND option. Note that the resulting count is in
characters, not code units. After a partial match, moving back from the
ovector[0] offset in the subject by the number of characters given for
the maximum lookbehind gets you to the earliest character that must be
retained. In a non-UTF or a 32-bit situation, moving back is just a
subtraction, but in UTF-8 or UTF-16 you have to count characters while
moving back through the code units.
Characters before the point you have now reached can be discarded, and
after the next segment has been added to what is retained, you should
run the next match with the startoffset argument set so that the match
begins at the same point as before.
For example, if the pattern "(?<=123)abc" is partially matched against
the string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maxi-
mum lookbehind count is 3, so all characters before offset 2 can be
discarded. The value of startoffset for the next match should be 3.
When pcre2test displays a partial match, it indicates the lookbehind
characters with '<' characters if the "allusedtext" modifier is set:
re> "(?<=123)abc"
data> xx123ab\=ph,allusedtext
Partial match: 123ab
<<< However, the "allusedtext" modifier is not avail-
able for JIT matching, because JIT matching does not maintain the first
and last consulted characters.
3. Matching a subject string that is split into multiple segments may
not always produce exactly the same result as matching over one single
long string when PCRE2_PARTIAL_SOFT is used. The section "Partial
Matching and Word Boundaries" above describes an issue that arises if
the pattern ends with \b or \B. Another kind of difference may occur
when there are multiple matching possibilities, because (for PCRE2_PAR-
TIAL_SOFT) a partial match result is given only when there are no com-
pleted matches. This means that as soon as the shortest match has been
found, continuation to a new subject segment is no longer possible.
Consider this pcre2test example:
re> /dog(sbody)?/
data> dogsb\=ps
0: dog
data> do\=ps,dfa
Partial match: do
data> gsb\=ps,dfa,dfa_restart
0: g
data> dogsbody\=dfa
0: dogsbody
1: dog
The first data line passes the string "dogsb" to a standard matching
function, setting the PCRE2_PARTIAL_SOFT option. Although the string is
a partial match for "dogsbody", the result is not PCRE2_ERROR_PARTIAL,
because the shorter string "dog" is a complete match. Similarly, when
the subject is presented to a DFA matching function in several parts
("do" and "gsb" being the first two) the match stops when "dog" has
been found, and it is not possible to continue. On the other hand, if
"dogsbody" is presented as a single string, a DFA matching function
finds both matches.
Because of these problems, it is best to use PCRE2_PARTIAL_HARD when
matching multi-segment data. The example above then behaves differ-
ently:
re> /dog(sbody)?/
data> dogsb\=ph
Partial match: dogsb
data> do\=ps,dfa
Partial match: do
data> gsb\=ph,dfa,dfa_restart
Partial match: gsb
4. Patterns that contain alternatives at the top level which do not all
start with the same pattern item may not work as expected when
PCRE2_DFA_RESTART is used. For example, consider this pattern:
to. This means that, for an unanchored pattern, if a continued match
fails, it is not possible to try again at a new starting point. All
this facility is capable of doing is continuing with the previous match
attempt. For example, consider this pattern:
1234|3789
@ -6037,29 +5992,16 @@ ISSUES WITH MULTI-SEGMENT MATCHING
the second alternative, because such a match does not start at the same
point in the subject string. Attempting to continue with the string
"7890" does not yield a match because only those alternatives that
match at one point in the subject are remembered. The problem arises
because the start of the second alternative matches within the first
alternative. There is no problem with anchored patterns or patterns
such as:
match at one point in the subject are remembered. Depending on the ap-
plication, this may or may not be what you want.
1234|ABCD
where no string can be a partial match for both alternatives. This is
not a problem if a standard matching function is used, because the en-
tire match has to be rerun each time:
re> /1234|3789/
data> ABC123\=ph
Partial match: 123
data> 1237890
0: 3789
Of course, instead of using PCRE2_DFA_RESTART, the same technique of
re-running the entire match can also be used with the DFA matching
function. Another possibility is to work with two buffers. If a partial
match at offset n in the first buffer is followed by "no match" when
PCRE2_DFA_RESTART is used on the second buffer, you can then try a new
match starting at offset n+1 in the first buffer.
If you do want to allow for starting again at the next character, one
way of doing it is to retain the matched part of the segment and try a
new complete match, as described for pcre2_match() above. Another pos-
sibility is to work with two buffers. If a partial match at offset n in
the first buffer is followed by "no match" when PCRE2_DFA_RESTART is
used on the second buffer, you can then try a new match starting at
offset n+1 in the first buffer.
AUTHOR
@ -6071,7 +6013,7 @@ AUTHOR
REVISION
Last updated: 22 July 2019
Last updated: 07 August 2019
Copyright (c) 1997-2019 University of Cambridge.
------------------------------------------------------------------------------

535
doc/pcre2partial.3
View File

@ -1,73 +1,107 @@
.TH PCRE2PARTIAL 3 "22 July 2019" "PCRE2 10.34"
.TH PCRE2PARTIAL 3 "07 August 2019" "PCRE2 10.34"
.SH NAME
PCRE2 - Perl-compatible regular expressions
.SH "PARTIAL MATCHING IN PCRE2"
.rs
.sp
In normal use of PCRE2, if the subject string that is passed to a matching
function matches as far as it goes, but is too short to match the entire
pattern, PCRE2_ERROR_NOMATCH is returned. There are circumstances where it
might be helpful to distinguish this case from other cases in which there is no
match.
In normal use of PCRE2, if there is a match up to the end of a subject string,
but more characters are needed to match the entire pattern, PCRE2_ERROR_NOMATCH
is returned, just like any other failing match. There are circumstances where
it might be helpful to distinguish this "partial match" case.
.P
Consider, for example, an application where a human is required to type in data
for a field with specific formatting requirements. An example might be a date
in the form \fIddmmmyy\fP, defined by this pattern:
.sp
^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$
.sp
If the application sees the user's keystrokes one by one, and can check that
what has been typed so far is potentially valid, it is able to raise an error
as soon as a mistake is made, by beeping and not reflecting the character that
has been typed, for example. This immediate feedback is likely to be a better
user interface than a check that is delayed until the entire string has been
entered. Partial matching can also be useful when the subject string is very
long and is not all available at once, as discussed below.
One example is an application where the subject string is very long, and not
all available at once. The requirement here is to be able to do the matching
segment by segment, but special action is needed when a matched substring spans
the boundary between two segments.
.P
PCRE2 supports partial matching by means of the PCRE2_PARTIAL_SOFT and
PCRE2_PARTIAL_HARD options, which can be set when calling a matching function.
The difference between the two options is whether or not a partial match is
preferred to an alternative complete match, though the details differ between
the two types of matching function. If both options are set, PCRE2_PARTIAL_HARD
takes precedence.
Another example is checking a user input string as it is typed, to ensure that
it conforms to a required format. Invalid characters can be immediately
diagnosed and rejected, giving instant feedback.
.P
If you want to use partial matching with just-in-time optimized code, you must
call \fBpcre2_jit_compile()\fP with one or both of these options:
Partial matching is a PCRE2-specific feature; it is not Perl-compatible. It is
requested by setting one of the PCRE2_PARTIAL_HARD or PCRE2_PARTIAL_SOFT
options when calling a matching function. The difference between the two
options is whether or not a partial match is preferred to an alternative
complete match, though the details differ between the two types of matching
function. If both options are set, PCRE2_PARTIAL_HARD takes precedence.
.P
If you want to use partial matching with just-in-time optimized code, as well
as setting a partial match option for the matching function, you must also call
\fBpcre2_jit_compile()\fP with one or both of these options:
.sp
PCRE2_JIT_PARTIAL_SOFT
PCRE2_JIT_PARTIAL_HARD
PCRE2_JIT_PARTIAL_SOFT
.sp
PCRE2_JIT_COMPLETE should also be set if you are going to run non-partial
matches on the same pattern. If the appropriate JIT mode has not been compiled,
interpretive matching code is used.
matches on the same pattern. Separate code is compiled for each mode. If the
appropriate JIT mode has not been compiled, interpretive matching code is used.
.P
Setting a partial matching option disables two of PCRE2's standard
optimizations. PCRE2 remembers the last literal code unit in a pattern, and
abandons matching immediately if it is not present in the subject string. This
optimization cannot be used for a subject string that might match only
partially. PCRE2 also knows the minimum length of a matching string, and does
optimization hints. PCRE2 remembers the last literal code unit in a pattern,
and abandons matching immediately if it is not present in the subject string.
This optimization cannot be used for a subject string that might match only
partially. PCRE2 also remembers a minimum length of a matching string, and does
not bother to run the matching function on shorter strings. This optimization
is also disabled for partial matching.
.
.
.SH "REQUIREMENTS FOR A PARTIAL MATCH"
.rs
.sp
A possible partial match occurs during matching when the end of the subject
string is reached successfully, but either more characters are needed to
complete the match, or the addition of more characters might change what is
matched.
.P
Example 1: if the pattern is /abc/ and the subject is "ab", more characters are
definitely needed to complete a match. In this case both hard and soft matching
options yield a partial match.
.P
Example 2: if the pattern is /ab+/ and the subject is "ab", a complete match
can be found, but the addition of more characters might change what is
matched. In this case, only PCRE2_PARTIAL_HARD returns a partial match;
PCRE2_PARTIAL_SOFT returns the complete match.
.P
On reaching the end of the subject, when PCRE2_PARTIAL_HARD is set, if the next
pattern item is \ez, \eZ, \eb, \eB, or $ there is always a partial match.
Otherwise, for both options, the next pattern item must be one that inspects a
character, and at least one of the following must be true:
.P
(1) At least one character has already been inspected. An inspected character
need not form part of the final matched string; lookbehind assertions and the
\eK escape sequence provide ways of inspecting characters before the start of a
matched string.
.P
(2) The pattern contains one or more lookbehind assertions. This condition
exists in case there is a lookbehind that inspects characters before the start
of the match.
.P
(3) There is a special case when the whole pattern can match an empty string.
When the starting point is at the end of the subject, the empty string match is
a possibility, and if PCRE2_PARTIAL_SOFT is set and neither of the above
conditions is true, it is returned. However, because adding more characters
might result in a non-empty match, PCRE2_PARTIAL_HARD returns a partial match,
which in this case means "there is going to be a match at this point, but until
some more characters are added, we do not know if it will be an empty string or
something longer".
.
.
.
.SH "PARTIAL MATCHING USING pcre2_match()"
.rs
.sp
A partial match occurs during a call to \fBpcre2_match()\fP when the end of the
subject string is reached successfully, but matching cannot continue because
more characters are needed, and in addition, either at least one character in
the subject has been inspected or the pattern contains a lookbehind, or (when
PCRE2_PARTIAL_HARD is set) the pattern could match an empty string. An
inspected character need not form part of the final matched string; lookbehind
assertions and the \eK escape sequence provide ways of inspecting characters
before the start of a matched string.
.P
The three additional requirements define the cases where adding more characters
to the existing subject may complete the same match that would occur if they
had all been present in the first place. Without these conditions there would
be a partial match of an empty string at the end of the subject for all
unanchored patterns (and also for anchored patterns if the subject itself is
empty).
When a partial matching option is set, the result of calling
\fBpcre2_match()\fP can be one of the following:
.TP 2
\fBA successful match\fP
A complete match has been found, starting and ending within this subject.
.TP
\fBPCRE2_ERROR_NOMATCH\fP
No match can start anywhere in this subject.
.TP
\fBPCRE2_ERROR_PARTIAL\fP
Adding more characters may result in a complete match that uses one or more
characters from the end of this subject.
.P
When a partial match is returned, the first two elements in the ovector point
to the portion of the subject that was matched, but the values in the rest of
@ -83,24 +117,6 @@ is "456abc12", a partial match is found for the string "abc12", because all
these characters are needed for a subsequent re-match with additional
characters.
.P
What happens when a partial match is identified depends on which of the two
partial matching options is set.
.
.
.SS "PCRE2_PARTIAL_SOFT WITH pcre2_match()"
.rs
.sp
If PCRE2_PARTIAL_SOFT is set when \fBpcre2_match()\fP identifies a partial
match, the partial match is remembered, but matching continues as normal, and
other alternatives in the pattern are tried. If no complete match can be found,
PCRE2_ERROR_PARTIAL is returned instead of PCRE2_ERROR_NOMATCH.
.P
This option is "soft" because it prefers a complete match over a partial match.
All the various matching items in a pattern behave as if the subject string is
potentially complete. For example, \ez, \eZ, and $ match at the end of the
subject, as normal, and for \eb and \eB the end of the subject is treated as a
non-alphanumeric.
.P
If there is more than one partial match, the first one that was found provides
the data that is returned. Consider this pattern:
.sp
@ -109,27 +125,32 @@ the data that is returned. Consider this pattern:
If this is matched against the subject string "abc123dog", both alternatives
fail to match, but the end of the subject is reached during matching, so
PCRE2_ERROR_PARTIAL is returned. The offsets are set to 3 and 9, identifying
"123dog" as the first partial match that was found. (In this example, there are
two partial matches, because "dog" on its own partially matches the second
alternative.)
"123dog" as the first partial match. (In this example, there are two partial
matches, because "dog" on its own partially matches the second alternative.)
.
.
.SS "PCRE2_PARTIAL_HARD WITH pcre2_match()"
.rs
.sp
If PCRE2_PARTIAL_HARD is set for \fBpcre2_match()\fP, PCRE2_ERROR_PARTIAL is
returned as soon as a partial match is found, without continuing to search for
possible complete matches. This option is "hard" because it prefers an earlier
partial match over a later complete match. For this reason, the assumption is
made that the end of the supplied subject string may not be the true end of the
available data, and so, if \ez, \eZ, \eb, \eB, or $ are encountered at the end
of the subject, the result is PCRE2_ERROR_PARTIAL, whether or not any
characters have been inspected.
.
.
.SS "Comparing hard and soft partial matching"
.SS "How a partial match is processed by pcre2_match()"
.rs
.sp
What happens when a partial match is identified depends on which of the two
partial matching options is set.
.P
If PCRE2_PARTIAL_HARD is set, PCRE2_ERROR_PARTIAL is returned as soon as a
partial match is found, without continuing to search for possible complete
matches. This option is "hard" because it prefers an earlier partial match over
a later complete match. For this reason, the assumption is made that the end of
the supplied subject string is not the true end of the available data, which is
why \ez, \eZ, \eb, \eB, and $ always give a partial match.
.P
If PCRE2_PARTIAL_SOFT is set, the partial match is remembered, but matching
continues as normal, and other alternatives in the pattern are tried. If no
complete match can be found, PCRE2_ERROR_PARTIAL is returned instead of
PCRE2_ERROR_NOMATCH. This option is "soft" because it prefers a complete match
over a partial match. All the various matching items in a pattern behave as if
the subject string is potentially complete; \ez, \eZ, and $ match at the end of
the subject, as normal, and for \eb and \eB the end of the subject is treated
as a non-alphanumeric.
.P
The difference between the two partial matching options can be illustrated by a
pattern such as:
.sp
@ -154,25 +175,129 @@ The second pattern will never match "dogsbody", because it will always find the
shorter match first.
.
.
.SS "Example of partial matching using pcre2test"
.rs
.sp
The \fBpcre2test\fP data modifiers \fBpartial_hard\fP (or \fBph\fP) and
\fBpartial_soft\fP (or \fBps\fP) set PCRE2_PARTIAL_HARD and PCRE2_PARTIAL_SOFT,
respectively, when calling \fBpcre2_match()\fP. Here is a run of
\fBpcre2test\fP using a pattern that matches the whole subject in the form of a
date:
.sp
re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
data> 25dec3\e=ph
Partial match: 23dec3
data> 3ju\e=ph
Partial match: 3ju
data> 3juj\e=ph
No match
.sp
This example gives the same results for both hard and soft partial matching
options. Here is an example where there is a difference:
.sp
re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
data> 25jun04\e=ps
0: 25jun04
1: jun
data> 25jun04\e=ph
Partial match: 25jun04
.sp
With PCRE2_PARTIAL_SOFT, the subject is matched completely. For
PCRE2_PARTIAL_HARD, however, the subject is assumed not to be complete, so
there is only a partial match.
.
.
.
.SH "MULTI-SEGMENT MATCHING WITH pcre2_match()"
.rs
.sp
PCRE was not originally designed with multi-segment matching in mind. However,
over time, features (including partial matching) that make multi-segment
matching possible have been added. The string is searched segment by segment by
calling \fBpcre2_match()\fP repeatedly, with the aim of achieving the same
results that would happen if the entire string was available for searching.
.P
Special logic must be implemented to handle a matched substring that spans a
segment boundary. PCRE2_PARTIAL_HARD should be used, because it returns a
partial match at the end of a segment whenever there is the possibility of
changing the match by adding more characters. The PCRE2_NOTBOL option should
also be set for all but the first segment.
.P
When a partial match occurs, the next segment must be added to the current
subject and the match re-run, using the \fIstartoffset\fP argument of
\fBpcre2_match()\fP to begin at the point where the partial match started.
Multi-segment matching is usually used to search for substrings in the middle
of very long sequences, so the patterns are normally not anchored. For example:
.sp
re> /\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed/
data> ...the date is 23ja\e=ph
Partial match: 23ja
data> ...the date is 23jan19 and on that day...\e=offset=15
0: 23jan19
1: jan
.sp
Note the use of the \fBoffset\fP modifier to start the new match where the
partial match was found.
.P
In this simple example, the next segment was just added to the one in which the
partial match was found. However, if there are memory constraints, it may be
necessary to discard text that precedes the partial match before adding the
next segment. In cases such as the above, where the pattern does not contain
any lookbehinds, it is sufficient to retain only the partially matched
substring. However, if a pattern contains a lookbehind assertion, characters
that precede the start of the partial match may have been inspected during the
matching process.
.P
The only lookbehind information that is available is the length of the longest
lookbehind in a pattern. This may not, of course, be at the start of the
pattern, but retaining that many characters before the partial match is
sufficient, if not always strictly necessary. The way to do this is as follows:
.P
Before doing any matching, find the length of the longest lookbehind in the
pattern by calling \fBpcre2_pattern_info()\fP with the PCRE2_INFO_MAXLOOKBEHIND
option. Note that the resulting count is in characters, not code units. After a
partial match, moving back from the ovector[0] offset in the subject by the
number of characters given for the maximum lookbehind gets you to the earliest
character that must be retained. In a non-UTF or a 32-bit situation, moving
back is just a subtraction, but in UTF-8 or UTF-16 you have to count characters
while moving back through the code units. Characters before the point you have
now reached can be discarded.
.P
For example, if the pattern "(?<=123)abc" is partially matched against the
string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maximum
lookbehind count is 3, so all characters before offset 2 can be discarded. The
value of \fBstartoffset\fP for the next match should be 3. When \fBpcre2test\fP
displays a partial match, it indicates the lookbehind characters with '<'
characters if the \fBallusedtext\fP modifier is set:
.sp
re> "(?<=123)abc"
data> xx123ab\e=ph,allusedtext
Partial match: 123ab
<<<
.sp
Note that the \fPallusedtext\fP modifier is not available for JIT matching,
because JIT matching does not maintain the first and last consulted characters.
.
.
.
.SH "PARTIAL MATCHING USING pcre2_dfa_match()"
.rs
.sp
The DFA functions move along the subject string character by character, without
The DFA function moves along the subject string character by character, without
backtracking, searching for all possible matches simultaneously. If the end of
the subject is reached before the end of the pattern, there is the possibility
of a partial match, again provided that at least one character has been
inspected.
of a partial match.
.P
When PCRE2_PARTIAL_SOFT is set, PCRE2_ERROR_PARTIAL is returned only if there
have been no complete matches. Otherwise, the complete matches are returned.
However, if PCRE2_PARTIAL_HARD is set, a partial match takes precedence over
any complete matches. The portion of the string that was matched when the
longest partial match was found is set as the first matching string.
If PCRE2_PARTIAL_HARD is set, a partial match takes precedence over any
complete matches. The portion of the string that was matched when the longest
partial match was found is set as the first matching string.
.P
Because the DFA functions always search for all possible matches, and there is
no difference between greedy and ungreedy repetition, their behaviour is
different from the standard functions when PCRE2_PARTIAL_HARD is set. Consider
the string "dog" matched against the ungreedy pattern shown above:
Because the DFA function always searches for all possible matches, and there is
no difference between greedy and ungreedy repetition, its behaviour is
different from the \fBpcre2_match()\fP. Consider the string "dog" matched
against this ungreedy pattern:
.sp
/dog(sbody)??/
.sp
@ -181,62 +306,17 @@ Whereas the standard function stops as soon as it finds the complete match for
returns that when PCRE2_PARTIAL_HARD is set.
.
.
.SH "PARTIAL MATCHING AND WORD BOUNDARIES"
.rs
.sp
If a pattern ends with one of sequences \eb or \eB, which test for word
boundaries, partial matching with PCRE2_PARTIAL_SOFT can give counter-intuitive
results. Consider this pattern:
.sp
/\ebcat\eb/
.sp
This matches "cat", provided there is a word boundary at either end. If the
subject string is "the cat", the comparison of the final "t" with a following
character cannot take place, so a partial match is found. However, normal
matching carries on, and \eb matches at the end of the subject when the last
character is a letter, so a complete match is found. The result, therefore, is
\fInot\fP PCRE2_ERROR_PARTIAL. Using PCRE2_PARTIAL_HARD in this case does yield
PCRE2_ERROR_PARTIAL, because then the partial match takes precedence.
.
.
.SH "EXAMPLE OF PARTIAL MATCHING USING PCRE2TEST"
.rs
.sp
If the \fBpartial_soft\fP (or \fBps\fP) modifier is present on a
\fBpcre2test\fP data line, the PCRE2_PARTIAL_SOFT option is used for the match.
Here is a run of \fBpcre2test\fP that uses the date example quoted above:
.sp
re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
data> 25jun04\e=ps
0: 25jun04
1: jun
data> 25dec3\e=ps
Partial match: 23dec3
data> 3ju\e=ps
Partial match: 3ju
data> 3juj\e=ps
No match
data> j\e=ps
No match
.sp
The first data string is matched completely, so \fBpcre2test\fP shows the
matched substrings. The remaining four strings do not match the complete
pattern, but the first two are partial matches. Similar output is obtained
if DFA matching is used.
.P
If the \fBpartial_hard\fP (or \fBph\fP) modifier is present on a
\fBpcre2test\fP data line, the PCRE2_PARTIAL_HARD option is set for the match.
.
.
.SH "MULTI-SEGMENT MATCHING WITH pcre2_dfa_match()"
.rs
.sp
When a partial match has been found using a DFA matching function, it is
When a partial match has been found using the DFA matching function, it is
possible to continue the match by providing additional subject data and calling
the function again with the same compiled regular expression, this time setting
the PCRE2_DFA_RESTART option. You must pass the same working space as before,
because this is where details of the previous partial match are stored. Here is
an example using \fBpcre2test\fP:
because this is where details of the previous partial match are stored. You can
set the PCRE2_PARTIAL_SOFT or PCRE2_PARTIAL_HARD options with PCRE2_DFA_RESTART
to continue partial matching over multiple segments. Here is an example using
\fBpcre2test\fP:
.sp
re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
data> 23ja\e=dfa,ps
@ -248,136 +328,10 @@ The first call has "23ja" as the subject, and requests partial matching; the
second call has "n05" as the subject for the continued (restarted) match.
Notice that when the match is complete, only the last part is shown; PCRE2 does
not retain the previously partially-matched string. It is up to the calling
program to do that if it needs to.
.P
That means that, for an unanchored pattern, if a continued match fails, it is
not possible to try again at a new starting point. All this facility is capable
of doing is continuing with the previous match attempt. In the previous
example, if the second set of data is "ug23" the result is no match, even
though there would be a match for "aug23" if the entire string were given at
once. Depending on the application, this may or may not be what you want.
The only way to allow for starting again at the next character is to retain the
matched part of the subject and try a new complete match.
.P
You can set the PCRE2_PARTIAL_SOFT or PCRE2_PARTIAL_HARD options with
PCRE2_DFA_RESTART to continue partial matching over multiple segments. This
facility can be used to pass very long subject strings to the DFA matching
functions.
.
.
.SH "MULTI-SEGMENT MATCHING WITH pcre2_match()"
.rs
.sp
Unlike the DFA function, it is not possible to restart the previous match with
a new segment of data when using \fBpcre2_match()\fP. Instead, new data must be
added to the previous subject string, and the entire match re-run, starting
from the point where the partial match occurred. Earlier data can be discarded.
.P
It is best to use PCRE2_PARTIAL_HARD in this situation, because it does not
treat the end of a segment as the end of the subject when matching \ez, \eZ,
\eb, \eB, and $. Consider an unanchored pattern that matches dates:
.sp
re> /\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed/
data> The date is 23ja\e=ph
Partial match: 23ja
.sp
At this stage, an application could discard the text preceding "23ja", add on
text from the next segment, and call the matching function again. Unlike the
DFA matching function, the entire matching string must always be available,
and the complete matching process occurs for each call, so more memory and more
processing time is needed.
.
.
.SH "ISSUES WITH MULTI-SEGMENT MATCHING"
.rs
.sp
Certain types of pattern may give problems with multi-segment matching,
whichever matching function is used.
.P
1. If the pattern contains a test for the beginning of a line, you need to pass
the PCRE2_NOTBOL option when the subject string for any call does start at the
beginning of a line. There is also a PCRE2_NOTEOL option, but in practice when
doing multi-segment matching you should be using PCRE2_PARTIAL_HARD, which
includes the effect of PCRE2_NOTEOL.
.P
2. If a pattern contains a lookbehind assertion, characters that precede the
start of the partial match may have been inspected during the matching process.
When using \fBpcre2_match()\fP, sufficient characters must be retained for the
next match attempt. You can ensure that enough characters are retained by doing
the following:
.P
Before doing any matching, find the length of the longest lookbehind in the
pattern by calling \fBpcre2_pattern_info()\fP with the PCRE2_INFO_MAXLOOKBEHIND
option. Note that the resulting count is in characters, not code units. After a
partial match, moving back from the ovector[0] offset in the subject by the
number of characters given for the maximum lookbehind gets you to the earliest
character that must be retained. In a non-UTF or a 32-bit situation, moving
back is just a subtraction, but in UTF-8 or UTF-16 you have to count characters
while moving back through the code units.
.P
Characters before the point you have now reached can be discarded, and after
the next segment has been added to what is retained, you should run the next
match with the \fBstartoffset\fP argument set so that the match begins at the
same point as before.
.P
For example, if the pattern "(?<=123)abc" is partially matched against the
string "xx123ab", the ovector offsets are 5 and 7 ("ab"). The maximum
lookbehind count is 3, so all characters before offset 2 can be discarded. The
value of \fBstartoffset\fP for the next match should be 3. When \fBpcre2test\fP
displays a partial match, it indicates the lookbehind characters with '<'
characters if the "allusedtext" modifier is set:
.sp
re> "(?<=123)abc"
data> xx123ab\e=ph,allusedtext
Partial match: 123ab
<<<
However, the "allusedtext" modifier is not available for JIT matching, because
JIT matching does not maintain the first and last consulted characters.
.P
3. Matching a subject string that is split into multiple segments may not
always produce exactly the same result as matching over one single long string
when PCRE2_PARTIAL_SOFT is used. The section "Partial Matching and Word
Boundaries" above describes an issue that arises if the pattern ends with \eb
or \eB. Another kind of difference may occur when there are multiple matching
possibilities, because (for PCRE2_PARTIAL_SOFT) a partial match result is given
only when there are no completed matches. This means that as soon as the
shortest match has been found, continuation to a new subject segment is no
longer possible. Consider this \fBpcre2test\fP example:
.sp
re> /dog(sbody)?/
data> dogsb\e=ps
0: dog
data> do\e=ps,dfa
Partial match: do
data> gsb\e=ps,dfa,dfa_restart
0: g
data> dogsbody\e=dfa
0: dogsbody
1: dog
.sp
The first data line passes the string "dogsb" to a standard matching function,
setting the PCRE2_PARTIAL_SOFT option. Although the string is a partial match
for "dogsbody", the result is not PCRE2_ERROR_PARTIAL, because the shorter
string "dog" is a complete match. Similarly, when the subject is presented to
a DFA matching function in several parts ("do" and "gsb" being the first two)
the match stops when "dog" has been found, and it is not possible to continue.
On the other hand, if "dogsbody" is presented as a single string, a DFA
matching function finds both matches.
.P
Because of these problems, it is best to use PCRE2_PARTIAL_HARD when matching
multi-segment data. The example above then behaves differently:
.sp
re> /dog(sbody)?/
data> dogsb\e=ph
Partial match: dogsb
data> do\e=ps,dfa
Partial match: do
data> gsb\e=ph,dfa,dfa_restart
Partial match: gsb
.sp
4. Patterns that contain alternatives at the top level which do not all start
with the same pattern item may not work as expected when PCRE2_DFA_RESTART is
used. For example, consider this pattern:
program to do that if it needs to. This means that, for an unanchored pattern,
if a continued match fails, it is not possible to try again at a new starting
point. All this facility is capable of doing is continuing with the previous
match attempt. For example, consider this pattern:
.sp
1234|3789
.sp
@ -386,28 +340,15 @@ alternative is found at offset 3. There is no partial match for the second
alternative, because such a match does not start at the same point in the
subject string. Attempting to continue with the string "7890" does not yield a
match because only those alternatives that match at one point in the subject
are remembered. The problem arises because the start of the second alternative
matches within the first alternative. There is no problem with anchored
patterns or patterns such as:
.sp
1234|ABCD
.sp
where no string can be a partial match for both alternatives. This is not a
problem if a standard matching function is used, because the entire match has
to be rerun each time:
.sp
re> /1234|3789/
data> ABC123\e=ph
Partial match: 123
data> 1237890
0: 3789
.sp
Of course, instead of using PCRE2_DFA_RESTART, the same technique of re-running
the entire match can also be used with the DFA matching function. Another
possibility is to work with two buffers. If a partial match at offset \fIn\fP
in the first buffer is followed by "no match" when PCRE2_DFA_RESTART is used on
the second buffer, you can then try a new match starting at offset \fIn+1\fP in
the first buffer.
are remembered. Depending on the application, this may or may not be what you
want.
.P
If you do want to allow for starting again at the next character, one way of
doing it is to retain the matched part of the segment and try a new complete
match, as described for \fBpcre2_match()\fP above. Another possibility is to
work with two buffers. If a partial match at offset \fIn\fP in the first buffer
is followed by "no match" when PCRE2_DFA_RESTART is used on the second buffer,
you can then try a new match starting at offset \fIn+1\fP in the first buffer.
.
.
.SH AUTHOR
@ -424,6 +365,6 @@ Cambridge, England.
.rs
.sp
.nf
Last updated: 22 July 2019
Last updated: 07 August 2019
Copyright (c) 1997-2019 University of Cambridge.
.fi