TagValidationTool#checkEntry do not have any logic to skip error checks other than no-translated entry.
Furhter investigation is impossible without reproduceable data and senario.
/***Checksentryforvalidtags.**@returnAn{@linkErrorReport}summarizingtheresults(willbeemptyif*noissuesfound)*/privateErrorReportcheckEntry(FileInfofi,SourceTextEntryste,TMXEntryte){ErrorReportreport=newErrorReport(ste,te);//ifthere's no translation, skip the string bugfix for://https://sourceforge.net/p/omegat/bugs/64/if(!te.isTranslated()||ste.getSrcText().isEmpty()){returnreport;}//Checkprintfvariablesif(Preferences.isPreference(Preferences.CHECK_ALL_PRINTF_TAGS)){TagValidation.inspectPrintfVariables(false,report);}elseif(Preferences.isPreference(Preferences.CHECK_SIMPLE_PRINTF_TAGS)){TagValidation.inspectPrintfVariables(true,report);}//ExtrachecksforPOfiles:if(fi.filterClass.getCanonicalName().equals("PoFilter")){TagValidation.inspectPOWhitespace(report);}TagValidation.inspectOmegaTTags(ste,report);if(Preferences.isPreference(Preferences.CHECK_JAVA_PATTERN_TAGS)){TagValidation.inspectJavaMessageFormat(report);}TagValidation.inspectRemovePattern(report);returnreport;}
Last edit: Hiroshi Miura 2023-07-03
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think the more consistent behavior would be for this option to be renamed to reflect that it deals with tags (its intention is rather obvious from the code). And then this option and the option in the Preferences > Tag Processing that prevents creating target documents with tag issues should temporarily disable any other checks selected in the Check Issues and cover only tags.
Both of these options are about files validity, not about QA.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What I’d like to know is whether the function was coded before we had all the other checks and then forgotten about, hence the inconsistency that we need to fix, or if it intended to leave them aside, in which case I'd be fine with temporarily reverting the labels.
I wrote that bug report because I wanted OmegaT to check for all the issues on leaving the segment and it did not do that.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What I’d like to know is whether the function was coded before we had all the other checks and then forgotten about, hence the inconsistency that we need to fix, or if it intended to leave them aside, in which case I'd be fine with temporarily reverting the labels.
When it was added, we had only tag validation. Everything else was (and still is) available in the QA script. Then tag validation was retired and superseded by the Check Issues which is designed to be extendable by scripts and plugins. When it was done, I think very little though was given to automatic tag validation which would happen when leaving a segment and creating target files. Tag validation is relatively simple task as far as computing goes, but even on medium-sized projects it might slow things down. Add to that compute-hungry terminology checks and LanguageTool, and it becomes a real annoyance if it pops up every now and then.
I wrote that bug report because I wanted OmegaT to check for all the issues on leaving the segment and it did not do that.
Fair enough, but I think with the current implementation of the Check Issues function with the scope always being an entire current document, it is not a very good idea. If somehow the check could be limited to the segment only, that would be quite helpful.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
TagValidationTool#checkEntry
do not have any logic to skip error checks other than no-translated entry.Furhter investigation is impossible without reproduceable data and senario.
Last edit: Hiroshi Miura 2023-07-03
I'll prepare something later. Thank you.
When no reproducible given, I'd like to close the issue.
Sorry, it's on my "todo" list. I'll be on it shortly.
See here:
https://sourceforge.net/p/omegat/bugs/1225/?limit=25#8cc4/2a02/1a77/7acf
The check I mention is the automatic check made when leaving a segment.
I think the more consistent behavior would be for this option to be renamed to reflect that it deals with tags (its intention is rather obvious from the code). And then this option and the option in the Preferences > Tag Processing that prevents creating target documents with tag issues should temporarily disable any other checks selected in the Check Issues and cover only tags.
Both of these options are about files validity, not about QA.
What I’d like to know is whether the function was coded before we had all the other checks and then forgotten about, hence the inconsistency that we need to fix, or if it intended to leave them aside, in which case I'd be fine with temporarily reverting the labels.
I wrote that bug report because I wanted OmegaT to check for all the issues on leaving the segment and it did not do that.
When it was added, we had only tag validation. Everything else was (and still is) available in the QA script. Then tag validation was retired and superseded by the Check Issues which is designed to be extendable by scripts and plugins. When it was done, I think very little though was given to automatic tag validation which would happen when leaving a segment and creating target files. Tag validation is relatively simple task as far as computing goes, but even on medium-sized projects it might slow things down. Add to that compute-hungry terminology checks and LanguageTool, and it becomes a real annoyance if it pops up every now and then.
Fair enough, but I think with the current implementation of the Check Issues function with the scope always being an entire current document, it is not a very good idea. If somehow the check could be limited to the segment only, that would be quite helpful.
I totally agree with you about limiting the scope to the current segment, in that particular case.