tasks: add FatcatDOI

1 files changed, 28 insertions, 2 deletions

diff --git a/python/refcat/tasks.py b/python/refcat/tasks.py
index 4f7c138..5553b23 100644
--- a/python/refcat/tasks.py
+++ b/python/refcat/tasks.py
@@ -346,7 +346,8 @@ class RefsDOI(Refcat):
 
 class RefsPMID(Refcat):
     """
-    Sorted (pmid, doc) tuples from refs; PMID is an integer.
+    Sorted (pmid, doc) tuples from refs; PMID is an integer,
+    https://www.ncbi.nlm.nih.gov/pmc/pmctopmid/
     """
     def requires(self):
         return RefsWithUnstructured()
@@ -369,7 +370,8 @@ class RefsPMID(Refcat):
 
 class RefsPMCID(Refcat):
     """
-    Sorted (pmcid, doc) tuples from refs, e.g. PMC2860560, ...
+    Sorted (pmcid, doc) tuples from refs, e.g. PMC2860560,
+    https://www.ncbi.nlm.nih.gov/pmc/pmctopmid/
     """
     def requires(self):
         return RefsWithUnstructured()
@@ -411,3 +413,27 @@ class RefsArxiv(Refcat):
 
     def output(self):
         return luigi.LocalTarget(path=self.path(ext="tsv.zst"), format=Zstd)
+
+
+class FatcatDOI(Refcat):
+    """
+    DOI from fatcat.
+    """
+    def requires(self):
+        return ReleaseExportReduced()
+
+    def run(self):
+        output = shellout("""
+                          zstdcat -T0 {input} |
+                          skate-map -m ff -x ext_ids.doi -skip-on-empty 1 |
+                          skate-cleanup -c doi -f 1 |
+                          LC_ALL=C sort -T {tmpdir} -k1,1 -S25% --parallel 4 |
+                          zstd -T0 -c > {output}
+                          """,
+                          n=self.n,
+                          tmpdir=self.tmpdir,
+                          input=self.input().path)
+        luigi.LocalTarget(output).move(self.output().path)
+
+    def output(self):
+        return luigi.LocalTarget(path=self.path(ext="tsv.zst"), format=Zstd)