From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-0.9 required=5.0 tests=BAYES_00,FORGED_GMAIL_RCVD, FREEMAIL_FROM autolearn=no autolearn_force=no version=3.4.4 X-Google-Thread: 103376,46b7978bb2005ec X-Google-NewGroupId: yes X-Google-Attributes: gida07f3367d7,domainid0,public,usenet X-Google-Language: ENGLISH,ASCII Path: g2news2.google.com!postnews.google.com!c19g2000yqe.googlegroups.com!not-for-mail From: Simon Belmont Newsgroups: comp.lang.ada Subject: Re: gprbuild with --db switch Date: Mon, 29 Aug 2011 14:28:42 -0700 (PDT) Organization: http://groups.google.com Message-ID: <28183cc0-e8ab-423c-a855-14951ea52376@c19g2000yqe.googlegroups.com> References: <9de55a2b-dea8-4351-b09f-77e6c3f79ba1@r12g2000vbe.googlegroups.com> NNTP-Posting-Host: 24.218.138.255 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Trace: posting.google.com 1314653322 18099 127.0.0.1 (29 Aug 2011 21:28:42 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Mon, 29 Aug 2011 21:28:42 +0000 (UTC) Complaints-To: groups-abuse@google.com Injection-Info: c19g2000yqe.googlegroups.com; posting-host=24.218.138.255; posting-account=ShYTIAoAAABytvcS76ZrG9GdaV-nXYKy User-Agent: G2/1.0 X-Google-Web-Client: true X-Google-Header-Order: ARLUEHNKC X-HTTP-UserAgent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.0; Trident/4.0; SLCC1; .NET CLR 2.0.50727; Media Center PC 5.0; .NET CLR 3.5.30729; .NET CLR 3.0.30729; .NET4.0C; InfoPath.2),gzip(gfe) Xref: g2news2.google.com comp.lang.ada:21697 Date: 2011-08-29T14:28:42-07:00 List-Id: On Aug 20, 1:42=A0pm, Simon Wright wrote: > Simon Belmont writes: > > Obstensibly, gprbuild will recognize a "--db dir" switch and pass it > > onto gprconfig when spawned (to add an additional knowledge base > > directory), but i'm having trouble getting it to do so. =A0Gprbuild wil= l > > not complain about the switch (and even spit out an error if the > > directory is invalid), but does not seems to invoke gprconfig with > > it. =A0By turning on the verbosity, the gprconfig line (as invoked by > > gprbuild) is quite clearly missing it, and consequently generates the > > auto.cgpr from only the original KB. =A0If i manually run the gprconfig > > line as generated by gprbuild, but add the --db switch, the > > configuration file is generated as intended. > > > Is this a known defenciency, or is there an order the switches must be > > in, or other 'gotcha' someone knows about? =A0I have tried most of the > > obvious things to no avail. =A0(Gprbuild Pro 1.4.1, 20100106, Linux) > > If you're supported (I see you're using Gprbuild Pro), this is the sort > of question AdaCore are only too happy to answer. > > If not, you could try the latest gprbuild (from GNAT GPL 2011) from > libre.adacore.com. You should be able to extract just gprbuild & friends > from the download. There may be configuration issues finding the > compiler (--target helps, ISTR), but you seem to be comfortable with > this sort of thing! Thank you for the feedback, but unfortunetly trying to switch versions would most likely be more paperwork hassle than simply finding a workaround, especially since putting the .xml in the original KB location (causes it to be correctly parsed). However, while I'm posting about silent gprbuild failures, I have another odd behavior that perhaps someone can verify, or at least point out what I'm doing incorrectly. Section 3.3 of the gprbuild users guide states "All attributes in a configuration project file are inherited by each user project file in the project tree", which i interpret to mean just that. For instance, if the .cgpr file includes a 'for Default_Language use "ada"', then any .gpr in the project can reference it, e.g. "foo :=3D project'Default_Language". The same dubious version of gprbuild, however, returns all these inherited attributes as a null string. Again, it accepts those inherited attributes as correct syntax (and reports errors if they do not exist), but they all return a null string. Does anyone know if this is this expected behavior? Again, thank you for time -sb