[Wave] Boost Inspection Report
Hi all, the boost inspection report claims that one of the Wave test files contains tabs. And this is correct! But: it needs to contain tabs because it's a testcase... Should I ignore this warning? Or is there a way to tell the inspect program to ignore tabs in this particular file? Regards Hartmut
Hartmut Kaiser wrote:
the boost inspection report claims that one of the Wave test files contains tabs. And this is correct!
But: it needs to contain tabs because it's a testcase...
Should I ignore this warning?
Yes.
Or is there a way to tell the inspect program to ignore tabs in this particular file?
Only way is to go change the code. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - Grafik/jabber.org
Rene Rivera wrote:
the boost inspection report claims that one of the Wave test files contains tabs. And this is correct!
But: it needs to contain tabs because it's a testcase...
Should I ignore this warning?
Yes.
Or is there a way to tell the inspect program to ignore tabs in this particular file?
Only way is to go change the code.
Will do. Do we have already a similar thing somewhere in the inspect tool (I'm asking to avoid inventing a new special syntax if there already exists one ;-)? Otherwise I'd go for adding the recognition of a special string somewhere in the file: BoostInspect:notab suppressing the -tab check for this particular file. Additionally this is extensible in the future, if needed. Comments? Regards Hartmut
Hartmut Kaiser wrote:
Rene Rivera wrote:
Only way is to go change the code.
Will do.
Do we have already a similar thing somewhere in the inspect tool (I'm asking to avoid inventing a new special syntax if there already exists one ;-)?
Not explicitly.. it's a very simple program :-)
Otherwise I'd go for adding the recognition of a special string somewhere in the file:
BoostInspect:notab
suppressing the -tab check for this particular file. Additionally this is extensible in the future, if needed.
Comments?
To take a cue from QuickBook... [boostinspect notab] But yes.. that's a good way of doing it. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - Grafik/jabber.org
Rene Rivera wrote:
Do we have already a similar thing somewhere in the inspect tool (I'm asking to avoid inventing a new special syntax if there already exists one ;-)?
Not explicitly.. it's a very simple program :-)
Otherwise I'd go for adding the recognition of a special string somewhere in the file:
BoostInspect:notab
suppressing the -tab check for this particular file. Additionally this is extensible in the future, if needed.
Comments?
To take a cue from QuickBook...
[boostinspect notab]
But yes.. that's a good way of doing it.
I added the checking for boostinspect:notab which may be easily added to any comments section in the target file. Regards Hartmut
Hartmut Kaiser wrote:
I added the checking for boostinspect:notab which may be easily added to any comments section in the target file.
Cool.. Just reran the test: 13 files with invalid line endings 0 bookmarks with invalid characters 0 invalid urls 170 broken links 40 unlinked files 21 file names too long 71 files with tabs 25 violations of the Boost min/max guidelines Bad thing is that many of the EOL problems are in headers.. Very bad as it causes numerous errors for some compilers! http://www.boost.org/regression-logs/inspection_report.html -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - Grafik/jabber.org
Rene Rivera wrote:
Hartmut Kaiser wrote:
I added the checking for boostinspect:notab which may be easily added to any comments section in the target file.
Cool.. Just reran the test:
13 files with invalid line endings 0 bookmarks with invalid characters 0 invalid urls 170 broken links 40 unlinked files 21 file names too long 71 files with tabs 25 violations of the Boost min/max guidelines
Bad thing is that many of the EOL problems are in headers.. Very bad as it causes numerous errors for some compilers!
Hi, spirit: libs/spirit/change_log.html: unlinked file But this is intentional. This is meant only as a stand alone documentation. What should I do about this? libs/spirit/test/escape_char_parser_tests.cpp: violation of Boost min/max guidelines on line 167, violation of Boost min/max guidelines on line 180 Sorry, what's the violation here? And where are the guidelines? assert(wc == std::numeric_limits<wchar_t>::max()); Thanks! -- Joel de Guzman http://www.boost-consulting.com http://spirit.sf.net
Joel de Guzman wrote:
Rene Rivera wrote:
Cool.. Just reran the test:
13 files with invalid line endings 0 bookmarks with invalid characters 0 invalid urls 170 broken links 40 unlinked files 21 file names too long 71 files with tabs 25 violations of the Boost min/max guidelines
Bad thing is that many of the EOL problems are in headers.. Very bad as it causes numerous errors for some compilers!
spirit:
libs/spirit/change_log.html: unlinked file
But this is intentional. This is meant only as a stand alone documentation. What should I do about this?
Two options: a) ignore the issue, after all these are only suggestions at the end; b) add a link to it from someplace in the spirit docs, after all how will users find the file otherwise?
libs/spirit/test/escape_char_parser_tests.cpp: violation of Boost min/max guidelines on line 167, violation of Boost min/max guidelines on line 180
Sorry, what's the violation here? And where are the guidelines?
assert(wc == std::numeric_limits<wchar_t>::max());
I think it would be: assert(wc == (std::numeric_limits<wchar_t>::max)()); It's one of the items in: http://www.boost.org/more/lib_guide.htm#Design_and_Programming -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - Grafik/jabber.org
Rene Rivera wrote:
Joel de Guzman wrote:
Rene Rivera wrote:
Cool.. Just reran the test:
13 files with invalid line endings 0 bookmarks with invalid characters 0 invalid urls 170 broken links 40 unlinked files 21 file names too long 71 files with tabs 25 violations of the Boost min/max guidelines
Bad thing is that many of the EOL problems are in headers.. Very bad as it causes numerous errors for some compilers!
spirit:
libs/spirit/change_log.html: unlinked file
But this is intentional. This is meant only as a stand alone documentation. What should I do about this?
Two options: a) ignore the issue, after all these are only suggestions at the end; b) add a link to it from someplace in the spirit docs, after all how will users find the file otherwise?
It's traditionally only a readme file for developers only. A snapshot of which is (again traditionally) linked in the Spirit front page at http://spirit.sf.net on each significant release. This is the same as boost's Latest News section where libraries list the changes per release.
libs/spirit/test/escape_char_parser_tests.cpp: violation of Boost min/max guidelines on line 167, violation of Boost min/max guidelines on line 180
Sorry, what's the violation here? And where are the guidelines?
assert(wc == std::numeric_limits<wchar_t>::max());
I think it would be:
assert(wc == (std::numeric_limits<wchar_t>::max)());
It's one of the items in:
http://www.boost.org/more/lib_guide.htm#Design_and_Programming
Darn! Forgetful me! I remember now doing some of the same tweaks once before. Thanks! Regards, -- Joel de Guzman http://www.boost-consulting.com http://spirit.sf.net
From: Hartmut Kaiser <hartmut.kaiser@gmail.com>
Do we have already a similar thing somewhere in the inspect tool (I'm asking to avoid inventing a new special syntax if there already exists one ;-)? Otherwise I'd go for adding the recognition of a special string somewhere in the file:
BoostInspect:notab
suppressing the -tab check for this particular file. Additionally this is extensible in the future, if needed.
Perhaps a begin/end delimiter pair to restrict the scope of the suppression? -- Rob Stewart stewart@sig.com Software Engineer http://www.sig.com Susquehanna International Group, LLP using std::disclaimer;
Rob Stewart wrote:
Do we have already a similar thing somewhere in the inspect tool (I'm asking to avoid inventing a new special syntax if there already exists one ;-)? Otherwise I'd go for adding the recognition of a special string somewhere in the file:
BoostInspect:notab
suppressing the -tab check for this particular file. Additionally this is extensible in the future, if needed.
Perhaps a begin/end delimiter pair to restrict the scope of the suppression?
All we have so far is one single simple use case, where tabs are used as input data for a test program. Your suggestion is nice, but overkill, IMHO. So I'm going to implement boostinspect:notab for tab analysis supression. If anybody wants to implement begin/end delimiters to restrict the scope of this, please go ahead. Regards Hartmut
participants (3)
-
Hartmut Kaiser -
Joel de Guzman -
Rene Rivera -
Rob Stewart