Merge pull request #29 from johnchase/issue-25

Issue 25

CHANGELOG.md

@@ -3,5 +3,7 @@
 ## Version 0.9.0 (changes since 0.9.0 release go here)
 
 ### Features
-* Bug fix: There was a small bug in the previous version that would occasionally cause issues when installing with conda
+* Added the ability to add a previously existing list of IDs to the create command  
 
+### Bugs
+* Bug fix: There was a small bug in the previous version that would occasionally cause issues when installing with conda

README.rst

@@ -63,6 +63,9 @@ Creating a list of IDs
 
    cual-id create ids 42 # writes 42 ids to stdout
    cual-id create ids 42 > my-ids.txt # writes 42 ids to my-ids.txt
+   cual-id create ids 42 --existing-ids my-ids.txt # creates ids that do not
+   # overlap with existing ids. Note this will only compare new ids to the first
+   # column of the file
 
 
 

cualid/fix.py

@@ -1,10 +1,13 @@
 from difflib import get_close_matches
 
 
-def fix_ids(correct_input, input_to_check, thresh=.5):
-    corr_ids = [e.strip().split('\t')[1] for e in correct_input]
+def parse_ids(input_file, col):
+    return [e.strip().split('\t')[col] for e in input_file]
+
 
-    broke_ids = [e.strip().split('\t')[0] for e in input_to_check]
+def fix_ids(correct_input, input_to_check, thresh=.5):
+    corr_ids = parse_ids(correct_input, 1)
+    broke_ids = parse_ids(input_to_check, 0)
 
     seen = set()
     for broke_id in broke_ids:

cualid/mint.py

@@ -16,12 +16,20 @@ def at_least_distance(query, existing, d=3):
     return True
 
 
-def create_ids(n, id_length, min_distance=3, failure_threshold=0.99):
+def create_ids(n, id_length,
+               min_distance=3,
+               failure_threshold=0.99,
+               existing_ids=None):
+    if existing_ids is not None:
+        hrids = existing_ids
+    else:
+        hrids = []
     uuids = []
-    hrids = []
+    initial_hrid_len = len(hrids)
     failures = 0
     trys = 1
-    while len(hrids) < n and failures/trys < failure_threshold:
+    while (len(hrids) - initial_hrid_len < n and
+           failures/trys < failure_threshold):
         trys += 1
         uuid_ = uuid.uuid4()
         hrid = uuid_.hex[-id_length:]

cualid/tests/test_fix.py

@@ -1,7 +1,7 @@
 import unittest
 import io
 
-from cualid.fix import fix_ids, format_output
+from cualid.fix import fix_ids, format_output, parse_ids
 
 
 class TestFormatOutput(unittest.TestCase):
@@ -60,6 +60,10 @@ def test_fix_ids_fix_all(self):
 
         self.assertTrue(tested)
 
+    def test_parse_ids(self):
+        exp = ['23bb9', 'c0cab', '87696', '7869b', 'd50cc']
+        self.assertEqual(parse_ids(identical, 0), exp)
+        self.assertEqual(parse_ids(correct1, 1), exp)
 
 correct1 = io.StringIO("c0b5d3ae-d2d4-4aa5-bd51-76c93e223bb9\t23bb9\n"
                        "0b95a10f-0610-434f-9734-4d2ac02c0cab\tc0cab\n"

scripts/cual-id

@@ -4,7 +4,7 @@ import sys
 
 from cualid.label import get_barcodes
 from cualid.mint import create_ids
-from cualid.fix import fix_ids, format_output
+from cualid.fix import fix_ids, format_output, parse_ids
 
 @click.group()
 def cli():
@@ -33,13 +33,17 @@ def barcode_generator(input, output_pdf, suppress_ids, barcode):
 @click.argument('number-of-ids', type=int)
 @click.option('-l', '--length', type=int, default=8, required=False)
 @click.option('-f', '--fail-threshold', type=float, default=0.99, required=False)
-def sample_id_generator(number_of_ids, length, fail_threshold):
+@click.option('-e', '--existing-ids', type=click.File('U'), default=None, required=False)
+def sample_id_generator(number_of_ids, length, fail_threshold, existing_ids):
     """This script generates a file of sample IDs"""
     if fail_threshold >= 1:
         raise ValueError("")
     collected = 0
+    if existing_ids:
+      existing_ids = parse_ids(existing_ids, 0)
     for uuid_, hrid in create_ids(number_of_ids, length,
-                                  failure_threshold=fail_threshold):
+                                  failure_threshold=fail_threshold,
+                                  existing_ids=existing_ids):
         collected += 1
         click.echo("%s\t%s" %(uuid_, hrid))