Mercurial > repos > brinkmanlab > make_unique_id

changeset 3:a2258ce2d58c draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/brinkmanlab/galaxy-tools/tree/master/make_unique_id commit 08b7b36e1eba516e4d7eb97086baf7ec8d177c5d"
author brinkmanlab
date Mon, 15 Jun 2020 19:32:32 -0400
parents c8bda09480ae
children f2656e644641
files make_unique_id.py
diffstat 1 files changed, 6 insertions(+), 4 deletions(-) [+]
line wrap: on
line diff
--- a/make_unique_id.py	Mon Jun 15 19:04:53 2020 -0400
+++ b/make_unique_id.py	Mon Jun 15 19:32:32 2020 -0400
@@ -29,14 +29,14 @@
     ids = defaultdict(int)
     
     def makeUnique(seq):
-        newid = seq.id[:16]  # Genbank has a max length for the id and sequence length number, truncate the sequence id if too long
+        seqlenlen = len(str(len(seq)))
+        newid = seq.id[:26 - seqlenlen]  # Genbank has a max length for the id and sequence length number, truncate the sequence id if too long
         count = ids[newid]
         ids[newid] += 1
         if count:
             suffix = "_" + str(count)
-            seqlenlen = len(str(len(seq)))
-            if len(newid) + len(suffix) + 1 + seqlenlen > 16:
-               newid = newid[:16 - seqlenlen - len(suffix)]
+            if len(newid) + len(suffix) + 1 + seqlenlen > 26:
+               newid = newid[:25 - seqlenlen - len(suffix)]
 
             newid += suffix
             seq.name += suffix
@@ -47,6 +47,7 @@
         seq.id = newid
         return seq
 
+
     paths = iter(sys.argv[2:])
 
     for input, output in zip(paths, paths):
@@ -56,3 +57,4 @@
             format
         )
 
+