Created attachment 143589 [details] shar file of new port framework The Basic Local Alignment Search Tool (BLAST) finds regions of local similarity between sequences. The program compares nucleotide or protein sequences to sequence databases and calculates the statistical significance of matches. BLAST can be used to infer functional and evolutionary relationships between sequences as well as help identify members of gene families.
Can you provide test logs? Poudriere? Redports? At least the output of "make check-plist" followed by "make stage-qa" ?
If you choose poudriere, make sure you use either "poudriere testport" or "poudriere bulk -t" in order to enable stage-qa checks, thanks!
This port was also built on a blank VM to verify depends (standard procedure for me). I'll check out poudriere and redports when I get a chance. <<<ROOT@freebsdports.acadix>>> /usr/wip/biology/ncbi-blast 47 # make check-plist ====> Checking for pkg-plist issues (check-plist) ===> Parsing plist ===> Checking for items in STAGEDIR missing from pkg-plist ===> Checking for directories owned by MTREEs ===> Checking for directories handled by dependencies ===> Checking for items in pkg-plist which are not in STAGEDIR ===> No pkg-plist issues found (check-plist) <<<ROOT@freebsdports.acadix>>> /usr/wip/biology/ncbi-blast 48 # make stage-qa ====> Running Q/A tests (stage-qa) <<<ROOT@freebsdports.acadix>>> /usr/wip/biology/ncbi-blast 49 #
thanks, please post when you have them. What you just provided is good enough to set this PR to "patch-ready" though.
major issue, it locks up during configuration phase: checking size of size_t... 8 checking for amq... /usr/sbin/amq checking whether /usr/sbin/amq -w works... This command (/usr/sbin/amq -w) never returns. The man page says that amd server must be running for -w option. If I kill the hung process, then it passes the configuration. I think you need to disable the check somehow. Is amq even needed for this port?
other than the amq thing, it passed all tests: =========================================================================== ====> Running Q/A tests (stage-qa) ====> Checking for pkg-plist issues (check-plist) ===> Parsing plist ===> Checking for items in STAGEDIR missing from pkg-plist ===> Checking for directories owned by MTREEs ===> Checking for directories handled by dependencies ===> Checking for items in pkg-plist which are not in STAGEDIR ===> No pkg-plist issues found (check-plist) ====>> Checking for staging violations... done
jason, did you get a chance to look at this?
No, 'been swamped lately, but I'll investigate ASAP.
Just pinging on this. It's been in my inbox for a while, at some point I'm going to throw the PR back on the heap again if it doesn't start moving.
alright, that's a total of 16 days with no response which is long enough, I think we'll all agree. Ping the PR when something changes, I'm throwing it back on the heap.
anything to report here?
Created attachment 148035 [details] New shar with amq -w patch Added patch to disable amq -w check, which reportedly hangs configure script on some systems.
There's a better way to do that which doesn't require a patch, e.g. CONFIGURE_ENV+= ncbi_cv_prog_amq_w=no in the makefile. Have you determine if there's any impact to building without amq? Or in other words, why is it's presence tested if it's not needed?
By the way, be careful generating shar files. I found no less than 3 extra files in there, including a log. Shar files should be only the port files. Everything else needs separate attachments. However, in the case of "todo" and "submitted", I don't think you wanted that at all.
There is more on this version of the configure patch than the previous version. this version defined __RUNETYPE_INTERNAL. Why? Both version remove help text about infodir and mandir. why? It seems that it doesn't do anything to me. I'm going to try building without a patch at all.
Damn. Problem 1, you used /LATEST in the MASTER_SITES definition, which is bad for obvious reasons. That is easily fixed. What's not fixable is that version 2.2.29 sources have been completely removed. The previous versions sources are there and the latest 2.2.30 sources are there, but not 2.2.29. It may have been removed intentionally. I'll try the next version but this is getting heavy very fast. (I've also fixed a few things again quietly. I did this the first time too and now I have to fix them same things again)
(In reply to John Marino from comment #15) > Both version remove help text about infodir and mandir. why? It seems that > it doesn't do anything to me. Now I know why. I'll use sed instead of a patch though.
I'll bring this back too although I wonder how FreeBSD 8 will look it. X ${REINPLACE_CMD} \ X -e "s|-D_THREAD_SAFE|-D_THREAD_SAFE -D__RUNETYPE_INTERNAL|g" \ X ${WRKSRC}/src/build-system/configure
fyi, your conversion of USE_PYTHON => USES=python was wrong, it overwrote the previous definition of USES.
(In reply to John Marino from comment #18) > I'll bring this back too although I wonder how FreeBSD 8 will look it. > > X ${REINPLACE_CMD} \ > X -e "s|-D_THREAD_SAFE|-D_THREAD_SAFE -D__RUNETYPE_INTERNAL|g" \ > X ${WRKSRC}/src/build-system/configure While ncbi-blast is building (takes a while, doesn't it?), I am wondering if this is actually legit. The message in the comment is typically seen when libraries from old system are mixed with new. Since you don't use poudriere, I'm wondering if you have a mixed jail. If this builds successfully, I'll have to rebuild it without this line to see if it's really needed. I suspect it's not and your system was/is dirty.
verdict: This definition is really needed.
A commit references this bug: Author: marino Date: Mon Oct 6 20:16:37 UTC 2014 New revision: 370220 URL: https://svnweb.freebsd.org/changeset/ports/370220 Log: Add new port biology/ncbi-blast PR: 190854 Submitted by: Jason Bacon The Basic Local Alignment Search Tool (BLAST) finds regions of local similarity between sequences. The program compares nucleotide or protein sequences to sequence databases and calculates the statistical significance of matches. BLAST can be used to infer functional and evolutionary relationships between sequences as well as help identify members of gene families. Changes: head/biology/Makefile head/biology/ncbi-blast/ head/biology/ncbi-blast/Makefile head/biology/ncbi-blast/distinfo head/biology/ncbi-blast/pkg-descr head/biology/ncbi-blast/pkg-plist
Okay, finishing thoughts: 1) I tested on F10/amd64 and F84/i386 with poudriere bulk -t, passed fine 2) I had to significantly updated the pkg-plist with the new version 2.2.30. Many files were added or switched locations 3) It was good that I got rid of the configure patch and replaced it with sed because it wouldn't applied to version 2.2.30. However with sed, it worked with the new version just fine. 4) I made a few tweaks that I didn't document. you should review at your leisure. This PR is finished now.