Re: [Doxygen-develop] 1.5.5 and Fortran
Brought to you by:
dimitri
From: Davide C. <dc...@ar...> - 2008-02-19 16:54:04
|
> Hi, > >> I try to use doxygen 1.5.5 (svn 612 ) with my fortran project. >> Unfortunately I get the following error >> >> Preprocessing C:/Users/Alin M Elena/tdtbuj/source/Constants.f90... >> Parsing file C:/Users/Alin M Elena/tdtbuj/source/Constants.f90... >> parse error in end >> <scopename>******************************************************************** >> Error in file C:/Users/Alin M Elena/tdtbuj/source/Constants.f90 line: 185, >> state: 4 > I tried SVN version several days ago and it did not work for me > either. Unfortunately, I do not have time to look into it. > > I think I'll have some time next week to fix this and look through > some issues in bugzilla. > > Oleg Hi Oleg and the others, I gave a try to the new version and I also got the same error with fortran, after some debugging I noticed that the problem is that constructs like ENDIF or READ(unit,END=20) are treated like END MODULE or END SUBROUTINE and so on; the following changes in fortranscanner.l RE's work for me: <InterfaceBody>"end"({BS_}"interface")?.* changed to <InterfaceBody>^{BS}"end"({BS_}"interface".*)? <Start,ModuleBody>"end"({BS_}(module|program))?.* changed to <Start,ModuleBody>^{BS}"end"({BS_}(module|program).*)?{BS}$ <SubprogBody>"end"({BS_}{SUBPROG})?.* changed to <SubprogBody>^{BS}"end"({BS_}{SUBPROG}.*)?{BS}$ But of course I am not a lex expert, so they may not be the correct ones, especially I do not understand how newline are managed in fortranscanner, so something may be wrong in that sense. Hope this helps, Davide -- __________________________________________________________ Davide Cesari ARPA-Servizio Idro Meteorologico __ tel (39) 051/525926 ||\ fax (39) 051/6497501 |||\ e-mail dc...@ar... |||/ www http://www.arpa.emr.it/sim --- Address: ARPA-SIM, Viale Silvani 6, 40122 Bologna, Italy __________________________________________________________ |