ascl/ads bibcode comparisons

ascl-ads-comparison/README.md

@@ -0,0 +1,24 @@
+much of this is ad-hoc, so there are a couple of manual steps required to reproduce 
+the intended result
+
+1. include a .ads\_key file with the text of your api key in the folder above this one
+
+2. the file ascl\_codes must be obtained through a wget or lynx download, with
+some manipulation to list every code line by line
+
+this is what I did:
+* edit /etc/lynx/lynx.cfg to have persistent cookies
+* log in to ascl.net/adm with lynx
+* lynx -width=999 -dump -nolist -nomargins http://ascl.net/code/utility/ascl2 |  awk -F: '{if (NF==2) { print "#",$0} else {print $0} }' > ascl2a.txt
+* sed -i s/\#//g ascl2a.txt
+* awk -F: '{print $2}' ascl2a.txt | awk '{printf("ascl.%s %s\n",$2,$3)}' > ascl2b.txt 
+* awk '{print $1}' ascl2b.txt  > ascl\_codes
+
+
+
+3. the file ads\_codes is obtained by running ads\_checker.py
+
+4. code\_comparison looks at ads\_codes and checks if each of the lines appears 
+in ascl\_codes
+
+

ascl-ads-comparison/ads_checker.py

@@ -0,0 +1,67 @@
+import sys
+import os
+import json
+import requests
+
+'''
+REQUIRES 1 FILE IN YOUR DIRECTORY:
+    - .ads_key      : text of your ads api key
+'''
+
+
+f = open(os.path.join(sys.path[0], "../.ads_key"), "r")
+key = f.read()
+#print("your key is: " + key + "\n")
+
+key = key.strip()
+
+#os.system("cat .ads_key")
+
+
+headers = {
+        "Authorization": "Bearer:" + key,
+}
+
+def check_pages(num):
+    return_list = []
+    curr_result = 0
+    for k in range(num):
+        #print("starting at resutlt " + str(curr_result))
+
+        params = (
+            ('q', 'ascl'),
+            ('fl',  'bibcode'),
+            ('rows', "3000"),
+            ('start', str(curr_result))
+        )
+
+        response = requests.get('https://api.adsabs.harvard.edu/v1/search/query', headers=headers, params=params)
+        #print(response)
+        data = response.json();
+        num_results = data["response"]["numFound"]
+
+        curr_result = curr_result + num_results
+
+        #print(str(num_results) + " total found on page " + str(k))
+
+        new_rel = []         
+        for i in data["response"]["docs"]:
+            bc = i['bibcode']
+            if "ascl.soft" in bc:
+                new_rel.append(bc)
+
+        #print(str(len(new_rel)) + " relevant found on page " + str(k))
+
+        return_list.extend(new_rel)
+    return_list.sort()
+    return return_list
+
+
+
+lst = check_pages(3)
+
+f = open(os.path.join(sys.path[0], "ads_codes"), "w+")
+
+for i in lst:
+    f.write(i + "\n")
+