New run file / genomes now successfully loaded into workflow.

chipathlon/genome.py

@@ -52,7 +52,7 @@ class Genome(object):
         return self.files.get("base_file")
 
     def get_chrom_sizes(self):
-        return self.files.get("chrom_sizes")
+        return self.files.get("chrom.sizes")
 
     def get_additional_files(self):
         return self.files.get("additional_files")
@@ -60,15 +60,18 @@ class Genome(object):
     def get_chr_fasta_files(self):
         return self.files.get("chr_fasta")
 
+    def get_all_files(self):
+        return [self.get_base_file(), self.get_chrom_sizes()] + self.get_additional_files() + self.get_chr_fasta_files()
+
     def _load_prefixes(self):
         self.file_prefix = "genome_%s_%s" % (self.assembly, self.tool)
         self.add_file_prefix = self.file_prefix if self.tool == "bowtie2" else self.file_prefix + os.path.splitext(self.base_file)[1]
         bowtie2_path_prefix, base_ext = os.path.splitext(self.base_file)
-        self.base_ext = base_ext
+        self.base_ext = base_ext[1:]
         self.path_prefix = bowtie2_path_prefix if self.tool == "bowtie2" else self.base_file
         return
 
-    def _add_file(self, name, path, gen_file_type="base_file"):
+    def _add_file(self, name, path, gen_file_type):
         """
         Adds a file to the internal files list.  Files are indexed
         by type.  There should only be a single base_file / chrom.sizes
@@ -123,4 +126,5 @@ class Genome(object):
             for f in files:
                 if f.startswith("chr"):
                     self._add_file("%s_%s" % (self.add_file_prefix, f), root + "/" + f, "chr_fasta")
+            break
         return

chipathlon/run_parser.py

@@ -34,7 +34,7 @@ class RunParser(object):
         """
         return len(self.errors) == 0 and not self.parse_error
 
-    def get_errors(self):
+    def get_error_string(self):
         """
         Returns the errors as a newline separated string.
         """
@@ -64,6 +64,9 @@ class RunParser(object):
                 return self.genomes[assembly][tool]
         return None
 
+    def get_genomes(self):
+        return [genome for gen_info in self.genomes.values() for genome in gen_info.values()]
+
     def _load_genomes(self):
         """
         Validate that the correct genomic information exists, we leave it up

chipathlon/workflow.py

@@ -13,6 +13,7 @@ import chipathlon.workflow_job
 import chipathlon.db
 import chipathlon.workflow_module
 import chipathlon.run_data
+from chipathlon.run_parser import RunParser
 from chipathlon.download_generator import DownloadGenerator
 from chipathlon.align_generator import AlignGenerator
 from chipathlon.remove_duplicates_generator import RemoveDuplicatesGenerator
@@ -21,7 +22,6 @@ import random
 from pprint import pprint
 from Pegasus.DAX3 import *
 
-
 class Workflow(object):
 
     def __init__(self, jobhome, run_file, param_file, config_file, host, username, password, rewrite=False, debug=False):
@@ -97,6 +97,11 @@ class Workflow(object):
         self.dax.addFile(f)
         return
 
+    def _add_genome_file(self, gen_file_obj):
+        self.files[gen_file_obj["name"]] = gen_file_obj["file"]
+        self.dax.addFile(gen_file_obj["file"])
+        return
+
     def _remove_job(self, name):
         """
             We don't delete out of master_jobs list so we can keep track of
@@ -174,9 +179,15 @@ class Workflow(object):
         return
 
     def _load_run_data(self):
-        self.run_data = chipathlon.run_data.RunData(self.run_file, self.mdb)
-        for genome_file in self.run_data.file_list:
-            self._add_file(genome_file["name"], genome_file["path"], genome_file["site"])
+        self.run_parser = RunParser(self.run_file, self.mdb)
+        if self.run_parser.is_valid():
+            for genome in self.run_parser.get_genomes():
+                for gen_file_obj in genome.get_all_files():
+                    self._add_genome_file(gen_file_obj)
+            self.runs = self.run_parser.get_runs()
+        else:
+            self.err += self.run_parser.get_error_string()
+        self._raise()
         return
 
     def _download_from_gridfs(self, gridfs_id, output_file):