Hi Michal, > I traced this down to the "wrap as in 2.9.2" option - with this on, a > new paragraph (in for example abstract) gets denoted by two empty lines > - even if the bibtex view in JabRef shows only one. > In this sense, I would consider it a bug - though the fix could be to > simply remove the option..
Can you send me the file you were using as original file (non wrapped 210 file or original file generated in 292 version)? If you were not using the file in the testbib. I never saw this problem at my end, unless the 210 file is already wrong structured. > No, I am not using Camel option. Anyway, 2.9.2 prints "@STRING", while > the 2.11dev (git) code prints "@String". Hmm, in my opinion, this probably a bug? Any one can confirm this is an expected behaviour or not? On Thu, May 22, 2014 at 12:39 PM, Michal Kaut <[email protected]> wrote: > Hello again, > > it looks like I have underestimated how big the changes are - looking at > it now, I agree that keeping the backwards compatibility does not make > much sense .. but thanks for trying :-) > > Below, I will nevertheless answer your comments, just in case you find > it useful: > >> >>> - new paragraph (in abstract) has an extra empty line >> I think you were using wrong work-flow or wrong file for the test. If >> you were using the file testbib/testjabref.bib whose structure is >> wrong, then the resulted saveas292 will be wrong as >> testbib/testjabref_210as292. >> bib. Because this file has a wrapped field >> but was wrapped in a wrong way, so the program will add extra \n\t >> into it. The correct work-flow will be open testjabref_292.bib in >> 2.11dev then save as v292 or save as v210. > > I traced this down to the "wrap as in 2.9.2" option - with this on, a > new paragraph (in for example abstract) gets denoted by two empty lines > - even if the bibtex view in JabRef shows only one. > In this sense, I would consider it a bug - though the fix could be to > simply remove the option.. > >> >> I never use @STRING field so I haven't test it. >> >>> - change in capitalization >> I did not see this, is this because you are still using camel option? > > No, I am not using Camel option. Anyway, 2.9.2 prints "@STRING", while > the 2.11dev (git) code prints "@String". > > > Regards, > Michal > > >> >> >>> In addition, I found a difference in sort order of some fields >> >> After 2.10 the BibtexEntryType.java is changed a lot. The optional >> fields and required fields are very different. If one need the field >> order is exactly the same as the 2.9.2, one should manually define the >> order. >> >> >> 2.10 also fixes some bugs in entry sorting. Some entries will be >> always different from 2.9.2. So I think backward-compatibility does >> not make sense any more. If one really want the bib file arranged in >> the same way as 2.9.2, he should just use 2.9.2. Upgrade all client to >> 2.10 makes more sense here. >> >> Keep the option to save as legacy database format will be very painful >> in the future, I think we should avoid to do so. >> >> I suggest a possible work flow for you: >> for each commit always use v292 or v210 (command line) to re-save the >> file. You can automate this by using git hook. >> >> On Tue, May 20, 2014 at 7:44 AM, Michal Kaut <[email protected]> wrote: >>> Once more, thanks a lot. (A small donation to the project is on the way.) >>> >>> I have tested it a bit and it almost works, i.e. the saved files are almost >>> identical to those from version 2.9. >>> >>> With both wrapping and sorting of fields switched to "as ver 2.9.2", I see >>> the >>> following minor differences when re-saving the file: >>> - new paragraph (in abstract) has an extra empty line, compared to v. 2.9. >>> - 2.9 has: LastWord\n\t\n\tFirstWord, 2.11 has: >>> LastWord\n\n\t\n\tFirstWord >>> - different format of "@STRING" definitions >>> - change in capitalization + extra padding in the new version >>> - this is not that important, as it affects only the header of the file >>> >>> In addition, I found a difference in sort order of some fields, but then I >>> realized that the fields were not supported for the entry type (pages for >>> BOOK, >>> etc) .. so can actually use the diff-file to find them and delete them from >>> the >>> database :-) >>> >>> >>> Cheers, >>> Michal >>> >>> >>> On 20.05.2014 15:27, Oliver Kopp wrote: >>>> Hi, >>>> >>>> I released a development snapshot at http://ge.tt/9Idc446?c >>>> >>>> The .zip is for MacOSX, the .jar is the file you want to test. The >>>> .exe is an installer. >>>> >>>> Cheers, >>>> >>>> Oliver >>>> >>>> >>>> 2014-05-20 9:26 GMT+02:00 Michal Kaut <[email protected]>: >>>>> Brilliant, thanks a lot! >>>>> >>>>> Unfortunately I do not have the build setup (and experience) to test the >>>>> feature >>>>> now, so I will wait until it comes to a released version - but from your >>>>> description it is exactly what I needed (and a bit extra) :-) >>>>> >>>>> Regards, >>>>> Michal >>>>> >>>>> >>>>> On 20.05.2014 00:47, JabrefW wrote: >>>>>> Hi Michal and Oliver, >>>>>> >>>>>> I have implemented the feature request #864. I also add codes to allow >>>>>> arbitrary order of fields which can be defined by the user. >>>>>> See Pull request #10 commit 898d972. I haven't been fully test it. >>>>>> Welcome to try. >>>>>> You can select in File-tab, at the bottom there are 3 radiobuttons: >>>>>> >>>>>> 1. "Sort fields in alphabeta order (as ver 2.10)" >>>>>> 2. "Sort fields in old fasion (as ver 2.9.2)" >>>>>> 3. "Save fields as user defined order" >>>>>> If the 3rd is selected, the fields in below textbox is going to be >>>>>> used as the order appeared in the string. Any fields are not in the >>>>>> string are going to be appended at the end with alphabeta order. >>>>>> >>>>>> BTW: to allow the version control work, one need to turn off the camel >>>>>> style also. >>>>>> >>>>>> Cheers, >>>>>> >>>>>> >>>>>> On Thu, May 15, 2014 at 6:47 AM, Michal Kaut <[email protected]> >>>>>> wrote: >>>>>>> Thanks again. >>>>>>> >>>>>>> I have created a feature request (nr. 864). >>>>>>> For the meantime, I have to hope that people won't discover 2.10 for a >>>>>>> while - I >>>>>>> had only one "case" so far, and managed to persuade him to downgrade .. >>>>>>> but I >>>>>>> realize this is not a sustainable strategy... >>>>>>> >>>>>>> >>>>>>> Michal >>>>>>> >>>>>>> >>>>>>> On 15.05.2014 14:36, Oliver Kopp wrote: >>>>>>>> Hi, >>>>>>>> >>>>>>>> 2014-05-15 14:18 GMT+02:00 Michal Kaut <[email protected]>: >>>>>>>> >>>>>>>>> Anyway, from what you are saying, it sounds like it does not make >>>>>>>>> much sense to >>>>>>>>> create a feature request for an additional tick-box under the field >>>>>>>>> saving options? >>>>>>>>> (Unfortunately, I have no experience in Java development, so I cannot >>>>>>>>> help myself.) >>>>>>>> >>>>>>>> It does make sense to file feature requests (and bug reports). That >>>>>>>> helps to track missing functionality. In case someone finds time, he >>>>>>>> directly knows where to start :) >>>>>>>> >>>>>>>> Regarding your feature request, maybe the original author of the patch >>>>>>>> for JabRef 2.10 can do something. But I didn't hear anything from him. >>>>>>>> Maybe, it's easier to push JabRef 2.10 throughout your department... >>>>>>>> >>>>>>>> Cheers, >>>>>>>> >>>>>>>> Oliver >>>>>>>> >>>>>>>> ------------------------------------------------------------------------------ >>>>>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >>>>>>>> Instantly run your Selenium tests across 300+ browser/OS combos. >>>>>>>> Get unparalleled scalability from the best Selenium testing platform >>>>>>>> available >>>>>>>> Simple to use. Nothing to install. Get started now for free." >>>>>>>> http://p.sf.net/sfu/SauceLabs >>>>>>>> _______________________________________________ >>>>>>>> Jabref-users mailing list >>>>>>>> [email protected] >>>>>>>> https://lists.sourceforge.net/lists/listinfo/jabref-users >>>>>>>> >>>>>>> >>>>>>> ------------------------------------------------------------------------------ >>>>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >>>>>>> Instantly run your Selenium tests across 300+ browser/OS combos. >>>>>>> Get unparalleled scalability from the best Selenium testing platform >>>>>>> available >>>>>>> Simple to use. Nothing to install. Get started now for free." >>>>>>> http://p.sf.net/sfu/SauceLabs >>>>>>> _______________________________________________ >>>>>>> Jabref-users mailing list >>>>>>> [email protected] >>>>>>> https://lists.sourceforge.net/lists/listinfo/jabref-users >>>>>> >>>>>> ------------------------------------------------------------------------------ >>>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >>>>>> Instantly run your Selenium tests across 300+ browser/OS combos. >>>>>> Get unparalleled scalability from the best Selenium testing platform >>>>>> available >>>>>> Simple to use. Nothing to install. Get started now for free." >>>>>> http://p.sf.net/sfu/SauceLabs >>>>>> _______________________________________________ >>>>>> Jabref-users mailing list >>>>>> [email protected] >>>>>> https://lists.sourceforge.net/lists/listinfo/jabref-users >>>>>> >>>>> >>>>> ------------------------------------------------------------------------------ >>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >>>>> Instantly run your Selenium tests across 300+ browser/OS combos. >>>>> Get unparalleled scalability from the best Selenium testing platform >>>>> available >>>>> Simple to use. Nothing to install. Get started now for free." >>>>> http://p.sf.net/sfu/SauceLabs >>>>> _______________________________________________ >>>>> Jabref-users mailing list >>>>> [email protected] >>>>> https://lists.sourceforge.net/lists/listinfo/jabref-users >>> >>> ------------------------------------------------------------------------------ >>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >>> Instantly run your Selenium tests across 300+ browser/OS combos. >>> Get unparalleled scalability from the best Selenium testing platform >>> available >>> Simple to use. Nothing to install. Get started now for free." >>> http://p.sf.net/sfu/SauceLabs >>> _______________________________________________ >>> Jabref-users mailing list >>> [email protected] >>> https://lists.sourceforge.net/lists/listinfo/jabref-users >> >> ------------------------------------------------------------------------------ >> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE >> Instantly run your Selenium tests across 300+ browser/OS combos. >> Get unparalleled scalability from the best Selenium testing platform >> available >> Simple to use. Nothing to install. Get started now for free." >> http://p.sf.net/sfu/SauceLabs >> _______________________________________________ >> Jabref-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/jabref-users >> > > ------------------------------------------------------------------------------ > "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE > Instantly run your Selenium tests across 300+ browser/OS combos. > Get unparalleled scalability from the best Selenium testing platform available > Simple to use. Nothing to install. Get started now for free." > http://p.sf.net/sfu/SauceLabs > _______________________________________________ > Jabref-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/jabref-users ------------------------------------------------------------------------------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free." http://p.sf.net/sfu/SauceLabs _______________________________________________ Jabref-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jabref-users
