non well formed numeric value about tripal_blast

[kyrenya]

working fine!!

[laceysanderson]

Try changing the ending on the file to txt and see if that works?

[laceysanderson]

Sorry @kyrenya, I don't have a fix yet as I've been having troubles replicating the problem. Which version of PHP are you using?

[laceysanderson]

Hi @kyrenya, I just pushed a commit fixing this issue. Can you test it and make sure it works for you? Your above solution may end up causing the alignments to not line up.

[kyrenya]

It works! Thx a lot! 👍

[laceysanderson]

This was introduced for Issue #54 where validation was not really happening... I'm surprised you're running into issues for this though as I just added extensive automated testing for it. I'll add your example to the tests and get back to you ASAP

[kyrenya]

Ok, I just removed completely the tripal_blast module, and git-cloned it again.
No warnings now, and FASTA validated correctly.
I think we can close this by now, I'll go on doing tests tomorrow, got some questions about 'recent jobs' thing and the files saved in the machine, but I will open other issue for that :)

Thanks a lot for your help and for fixing the problem!!

[laceysanderson]

Hmmm.... this seems to be related to the query/hit coordinates. Could you provide the BLAST XML for this result so I can debug the problem?

[kyrenya]

sure!
it happens everytime with all the tests I've done.
uhmm, I cannot upload it, xml is not supported.. alternative?
email?

[kyrenya]

done! :)
2019Feb07_141328.blast.txt

[kyrenya]

hi! any update in this? is it anyway possible to avoid the warnings show for some specific role? (i.e. anonymous users) Thanks!

[kyrenya]

hi There! php 7.2!

[kyrenya]

so, I know nothing about php, but I'm trying to figure out what is happening with the lines of code regarding to the warnings (92 and 106 of blast_report_alignment_row.tpl.php), trying to figure out is the problem is actually with that script.

if I change the line:
$coord['qstart'] = ($k == 0) ? $q_from : $coord['qstop'] + 1;

for sth like:
$coord['qstart'] = ($k == 0) ? $q_from : 777;

of course in the alignment displays, my query frame start with 777 which is not the ideal thing, but warning about line 92 is removed now, getting only the warning about line 106.
But I have no idea of how to fix the code, in case is this the problem.
Is this helpful at all ? Can you shed some light on it despite you not being able to reproduce the problem?

Thanks a lot in advance.

[kyrenya]

after some googling,
I added this to the code:

settype($coord['qstart'], "integer");
settype($coord['qstop'], "integer");
settype($coord['hstart'], "integer");
settype($coord['hstop'], "integer");

apparently everything is fine now..

[laceysanderson]

Great! 🎉

[kyrenya]

uhmm, other error arised, it doesnt seem to recognise now the FASTA sequence when its a protein..
I attach screenshot

[kyrenya]

it's weird because before the 'git pull' it worked fine... :S
Thanks for taking care!

[laceysanderson]

That header format with my own example protein passes validation perfectly. Can you provide me with your protein sequence for easy testing?

[kyrenya]

I'm using the one that tripal_blast loads as example by default...

gi|166477|gb|AAA96434.1| resveratrol synthase [Arachis hypogaea]
MVSVSGIRKVQRAEGPATVLAIGTANPPNCIDQSTYADYYFRVTNSEHMTDLKKKFQRICERTQIKNRHM
YLTEEILKENPNMCAYKAPSLDAREDMMIREVPRVGKEAATKAIKEWGQPMSKITHLIFCTTSGVALPGV
DYELIVLLGLDPCVKRYMMYHQGCFAGGTVLRLAKDLAENNKDARVLIVCSENTAVTFRGPSETDMDSLV
GQALFADGAAAIIIGSDPVPEVEKPIFELVSTDQKLVPGSHGAIGGLLREVGLTFYLNKSVPDIISQNIN
DALNKAFDPLGISDYNSIFWIAHPGGRAILDQVEQKVNLKPEKMKATRDVLSNYGNMSSACVFFIMDLMR
KRSLEEGLKTTGEGLDWGVLFGFGPGLTIETVVLRSVAI

I tried with anything, even sth like

test
PP

[kyrenya]

the '>' was not pasted in my message, but it is there :)

[laceysanderson]

I think I found the bug in the regex! I pushed a commit to the main branch and that sequence now works for me. Can you test on your end as well?

[kyrenya]

it worked!!
but I got again the non well formed numeric value. warning.. shouldnt be fixed if I'm git pull-ing the tripal_blast thingy?