add MAG importer; reorder directory class listing

2 files changed, 73 insertions, 10 deletions

diff --git a/chocula/directories/__init__.py b/chocula/directories/__init__.py
index 17329e1..ed306c0 100644
--- a/chocula/directories/__init__.py
+++ b/chocula/directories/__init__.py
@@ -18,26 +18,29 @@ from chocula.directories.vanished_inactive import VanishedInactiveLoader
 from chocula.directories.issn_meta import IssnMetaLoader
 from chocula.directories.australian_era import AustralianEraLoader
 from chocula.directories.awol import AwolLoader
+from chocula.directories.mag import MagLoader
 
+# sort order roughly results in metadata prioritization
 ALL_CHOCULA_DIR_CLASSES = [
-    CrossrefLoader,
+    IssnMetaLoader,
+    ManualHomepageLoader,
+    ScieloLoader,
     DoajLoader,
+    CrossrefLoader,
     EntrezLoader,
     EzbLoader,
     GoldOALoader,
     NorwegianLoader,
+    AustralianEraLoader,
+    SzczepanskiLoader,
+    WikidataLoader,
+    AwolLoader,
+    VanishedDisapearedLoader,
+    VanishedInactiveLoader,
     OpenAPCLoader,
     RoadLoader,
     SherpaRomeoLoader,
-    SzczepanskiLoader,
-    WikidataLoader,
     SimLoader,
-    ScieloLoader,
-    ManualHomepageLoader,
     ZdbFizeLoader,
-    VanishedDisapearedLoader,
-    VanishedInactiveLoader,
-    IssnMetaLoader,
-    AustralianEraLoader,
-    AwolLoader,
+    MagLoader,
 ]
diff --git a/chocula/directories/mag.py b/chocula/directories/mag.py
new file mode 100644
index 0000000..dbbc324
--- /dev/null
+++ b/chocula/directories/mag.py
@@ -0,0 +1,60 @@
+from typing import Iterable, Optional
+import csv
+
+from chocula.util import clean_str, clean_issn
+from chocula.common import DirectoryLoader
+from chocula.database import DirectoryInfo, HomepageUrl
+
+
+class MagLoader(DirectoryLoader):
+    """
+    TSV Columns (from schema docs):
+
+        1	JournalId	long	PRIMARY KEY
+        2	Rank	uint	See FAQ
+        3	NormalizedName	string	
+        4	DisplayName	string	
+        5	Issn	string	
+        6	Publisher	string	
+        7	Webpage	string	
+        8	PaperCount	long	
+        9	PaperFamilyCount	long	See FAQ
+        10	CitationCount	long	
+        11	CreatedDate	DateTime
+
+    """
+
+    source_slug = "mag"
+
+    def open_file(self) -> Iterable:
+        return csv.DictReader(
+            open(self.config.mag.filepath, "r"),
+            delimiter="\t",
+            fieldnames=[
+                "JournalId",
+                "Rank",
+                "NormalizedName",
+                "DisplayName",
+                "Issn",
+                "Publisher",
+                "Webpage",
+                "PaperCount",
+                "PaperFamilyCount",
+                "CitationCount",
+                "CreatedDate",
+            ],
+        )
+
+    def parse_record(self, record) -> Optional[DirectoryInfo]:
+        info = DirectoryInfo(
+            directory_slug=self.source_slug,
+            raw_issn=clean_issn(record["Issn"]),
+            custom_id=record["JournalId"],
+            name=clean_str(record["DisplayName"]),
+            publisher=clean_str(record["Publisher"]),
+        )
+        homepage = HomepageUrl.from_url(record["Webpage"] or "")
+        if homepage:
+            info.homepage_urls.append(homepage)
+
+        return info