From a6984cfa2e80ef3f70255e3e26c5954210ef2dac Mon Sep 17 00:00:00 2001
From: Diane Trout <diane@caltech.edu>
Date: Tue, 21 Dec 2010 15:53:59 -0800
Subject: [PATCH] Add utility to fix the off-by-3 error in the amplified from
 field.

The bug appeared because the library name and library primary key
drifted apart because of some accidental deletions.

I eventually moved us back to using raw primary keys as that
made constructing django urls simplier & easier. However
I'd apparently forgotten to adjust the amplified from field.
---
 extra/fix_ob3.py | 77 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 77 insertions(+)
 create mode 100644 extra/fix_ob3.py

diff --git a/extra/fix_ob3.py b/extra/fix_ob3.py
new file mode 100644
index 0000000..1278901
--- /dev/null
+++ b/extra/fix_ob3.py
@@ -0,0 +1,77 @@
+"""
+Fix off by 3 error in our database
+"""
+from optparse import OptionParser
+import os
+import re
+import sys
+
+from django.core.management import setup_environ
+from htsworkflow.frontend import settings
+setup_environ(settings)
+
+import htsworkflow.frontend.samples.models as samples
+
+def main(cmdline=None):
+    parser = make_parser()
+    opts, args = parser.parse_args(cmdline)
+
+    dry_run = not opts.run
+    fix_ob3(dry_run=dry_run)
+
+    return 0
+
+def make_parser():
+    parser = OptionParser("%prog: fix off by 3 error that creeped in")
+    parser.add_option("--run", default=False, action="store_true",
+                      help="change the database")
+    return parser
+
+def fix_ob3(dry_run=True):
+    libraries = samples.Library.objects.order_by('id')
+
+    mismatch = 0
+    wrong_amp = 0
+    wrong_amp_ids = set()
+    for lib in libraries:
+        if lib.amplified_from_sample is not None:
+            amp_sample = lib.amplified_from_sample
+            alt_sample = samples.Library.objects.get(pk=int(amp_sample.id)-3)
+            if is_alt_sample_right(lib, alt_sample):
+                wrong_amp += 1
+                wrong_amp_ids.add(int(lib.id))
+                print "--- wrong lib ---"
+                display_names(lib, amp_sample, alt_sample)
+                if not dry_run:
+                    lib.amplified_from_sample = alt_sample
+                    lib.save()
+            #elif lib_name != amp_sample.library_name:
+            #    mismatch += 1
+            #    print "--- didn't match ---"
+            #    display_names(lib_name, lib, amp_sample, other_sample)
+    print "-----"
+    print "{0} mismatches".format(mismatch)
+    print "{0} obviously wrong libs".format(wrong_amp)
+    if len(wrong_amp_ids) > 0:
+        print "    {0} - {1}".format(min(wrong_amp_ids), max(wrong_amp_ids))
+
+def clean_lib_name(lib):
+    """Strip trailing amplified marker character"""
+    return re.sub(" *a$", "", lib.library_name)
+
+def is_alt_sample_right(lib, alt_sample):
+    """Check to see if the alt sample is the right sample
+    """
+    lib_name = clean_lib_name(lib)
+    return lib_name == alt_sample.library_name
+    
+def display_names(lib, amp_from_sample, alt_from_sample):
+    lib_name = clean_lib_name(lib)
+    print "NonA: "+lib.id+"|"+lib_name+"|"
+    print "   A: "+lib.id+"|"+lib.library_name+"|"
+    print "AmpF: "+ amp_from_sample.id+"|"+amp_from_sample.library_name+"|"
+    print "FixF: "+ alt_from_sample.id+"|"+alt_from_sample.library_name+"|"
+
+
+if __name__ == "__main__":
+    sys.exit(main())
-- 
2.30.2