Mercurial > repos > drosofff > fetch_fasta_from_ncbi

changeset 1:c1d17d173128 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author drosofff
date Tue, 12 May 2015 17:42:29 -0400
parents 4b34f2b5c14e
children e9df554f7725
files retrieve_fasta_from_NCBI.py
diffstat 1 files changed, 6 insertions(+), 2 deletions(-) [+]
line wrap: on
line diff
--- a/retrieve_fasta_from_NCBI.py	Mon Apr 13 18:17:08 2015 -0400
+++ b/retrieve_fasta_from_NCBI.py	Tue May 12 17:42:29 2015 -0400
@@ -150,6 +150,8 @@
         req = urllib2.Request(url, data)
         response = urllib2.urlopen(req)
         fasta = response.read()
+        if "Resource temporarily unavailable" in fasta:
+            return '' # to reiterate the failed download
         if self.dbname != "pubmed":
             assert fasta.startswith(">"), fasta
         fasta = self.sanitiser(self.dbname, fasta) #
@@ -212,8 +214,10 @@
                 end = min(count, start+batch_size)
                 batch = uids_list[start:end]
                 self.epost(self.dbname, ",".join(batch))
-                self.logger.info("retrieving batch %d" % ((start / batch_size) + 1))
-                mfasta = self.efetch(self.dbname, self.query_key, self.webenv)
+                mfasta = ''
+                while not mfasta:
+                    self.logger.info("retrieving batch %d" % ((start / batch_size) + 1))
+                    mfasta = self.efetch(self.dbname, self.query_key, self.webenv)
                 out.write(mfasta + '\n')