Windows 10, TeXLive 2016, and grep
Where can I find the applicable documentation for the grep script included in TeXLive on Windows? I.e. the one called with `mtxrun -script grep ...`. In particular I'm interested in knowing which grep options are available and which are not. `man mtx-grep` lists only 4 options (pattern, count, nocomment, and xml). Are those really the only options available? -- ✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝ Br. Samuel, OSB St. Anselm’s Abbey Washington, DC (R. Padraic Springuel) PAX ☧ ΧΡΙΣΤΟΣ
On 8/16/2016 9:52 PM, Br. Samuel Springuel wrote:
Where can I find the applicable documentation for the grep script included in TeXLive on Windows? I.e. the one called with `mtxrun -script grep ...`. In particular I'm interested in knowing which grep options are available and which are not. `man mtx-grep` lists only 4 options (pattern, count, nocomment, and xml). Are those really the only options available?
well, we can add more if needed .. i mostly wrote it for searching tex files (using the lua pattern syntax) as fast as possible Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------
I'm looking particularly for the `-m #` (return only first # matches) and `-o` (return only the portion of the line which matches), with the latter being more important than the former. I have a line in a Unix script (where I can expect standard grep to be available) which looks like this: grep -m 1 'GREGORIO_VERSION' $loc | grep -o '[0-9]*\.[0-9]*\.[0-9]*-*[betarc]*[0-9]*' >> $OUTPUT 2>&1 Basically the script is supposed to identify which version of a file in our package is a person has. To do this it looks for the first line in a file ($loc) which is marked as containing the version number (by looking for the text of the comment which we use to mark those lines for scripting purposes) and then extract the version number from that line (by looking for the pattern which the version number follows). In Windows, I can do the first part of this command with `findstr` (a native command which has similar, but not identical, functionality to grep, but cannot do the second at all. My current work around is to use the characters which precede and follow the version number to trim the string, but this requires me to have repetitive code because not all the files in our package use the same delimiting characters. What I accomplish in one `for` loop in the Unix version (looping over a list of the files in our package), I need three loops to do in Windows (breaking the list up by the delimiting characters). Having `mtxrun -script grep` support the `-o` option would enable me to consolidate and simplify the code. -- ✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝✝ Br. Samuel, OSB St. Anselm’s Abbey Washington, DC (R. Padraic Springuel) PAX ☧ ΧΡΙΣΤΟΣ
On 8/20/2016 3:08 AM, Br. Samuel Springuel wrote:
I'm looking particularly for the `-m #` (return only first # matches) and `-o` (return only the portion of the line which matches), with the latter being more important than the former.
I have a line in a Unix script (where I can expect standard grep to be available) which looks like this:
grep -m 1 'GREGORIO_VERSION' $loc | grep -o '[0-9]*\.[0-9]*\.[0-9]*-*[betarc]*[0-9]*' >> $OUTPUT 2>&1
Basically the script is supposed to identify which version of a file in our package is a person has. To do this it looks for the first line in a file ($loc) which is marked as containing the version number (by looking for the text of the comment which we use to mark those lines for scripting purposes) and then extract the version number from that line (by looking for the pattern which the version number follows).
In Windows, I can do the first part of this command with `findstr` (a native command which has similar, but not identical, functionality to grep, but cannot do the second at all. My current work around is to use the characters which precede and follow the version number to trim the string, but this requires me to have repetitive code because not all the files in our package use the same delimiting characters. What I accomplish in one `for` loop in the Unix version (looping over a list of the files in our package), I need three loops to do in Windows (breaking the list up by the delimiting characters). Having `mtxrun -script grep` support the `-o` option would enable me to consolidate and simplify the code.
I'll add -n=1 (--first) and --match
rgrep --pattern="modules.-%['(.-)'%]" char-*.lua --first --match
which returns (on my machine) char-act char-cjk char-def char-enc char-ent char-fio char-ini char-map char-rep char-tex char-utf ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------
participants (2)
-
Br. Samuel Springuel -
Hans Hagen