start a Makefile

Move all "index" functions into classes, each in a separate file.

Add lots of type annotations.

Use dataclass objects to hold database rows. This aspect will need
further refactoring to remove "extra" usage, probably by adding database
rows to align with DatabaseInfo more closely.

1 files changed, 36 insertions, 0 deletions

diff --git a/chocula/directories/crossref.py b/chocula/directories/crossref.py
new file mode 100644
index 0000000..ba47566
--- /dev/null
+++ b/chocula/directories/crossref.py
@@ -0,0 +1,36 @@
+
+from typing import Iterable, Optional
+import csv
+
+from chocula.util import clean_str
+from chocula.common import DirectoryLoader
+from chocula.database import DirectoryInfo
+
+
+class CrossrefLoader(DirectoryLoader):
+    """
+    CSV Columns:
+
+        #"JournalTitle","JournalID","Publisher","pissn","eissn","additionalIssns","doi","(year1)[volume1]issue1,issue2,issue3(year2)[volume2]issue4,issues5"
+
+    """
+
+    source_slug = "crossref"
+
+    def open_file(self) -> Iterable:
+        return csv.DictReader(open(self.config.CROSSREF_FILE))
+
+    def parse_record(self, record) -> Optional[DirectoryInfo]:
+        info = DirectoryInfo(
+            directory_slug=self.source_slug,
+            issne=record['eissn'],
+            issnp=record['pissn'],
+            custom_id=record.get('doi').strip() or None,
+            name=clean_str(record.get('JournalTitle')),
+            publisher=clean_str(record.get('Publisher')),
+        )
+
+        if record['additionalIssns']:
+            info.raw_issn = record['additionalIssns'][0]
+
+        return info