Changeset 566 for trunk/Update Dates/update_dates.py

Timestamp:

07/19/09 22:32:55 (15 years ago)

Author:

Nicholas Riley

Message:

update_dates.py: Continuous scanning workflow.

File:

• trunk/Update Dates/update_dates.py (modified) (3 diffs)

trunk/Update Dates/update_dates.py

@@ -1 +1 @@
 from appscript import *
+from datetime import datetime
+from osax import *
+from plistlib import readPlist, writePlist
+import os
 import re
+import time
+
+PREFERENCES_PATH = \
+    os.path.expanduser('~/Library/Preferences/net.sabi.UpdateDates.plist')
 
 DATE_FORMATS = (('%m/%d/%y',  r'\d{1,2}/\d{1,2}/\d{1,2}'       ), # T-Mobile
@@ -30 +38 @@
         matched = m.group(matched_format).replace(' ', '')
         try:
-            parsed = datetime.datetime.strptime(matched, format)
+            parsed = datetime.strptime(matched, format)
         except ValueError, e: # not a date
             no_match.append((matched, format, e))
@@ -39 +47 @@
     return None, no_match
 
+def extract_source(title, hint):
+    if hint:
+        return title[:RE_DATE.search(title).start(0)].rstrip()
+    else:
+        return title
+
 EagleFiler = app(id='com.c-command.EagleFiler')
 Paper = EagleFiler.documents['Paper.eflibrary']
 
-for record in Paper.library_records[its.kind=='PDF']():
-    title = record.title()
-    hint, no_match = extract_date(title)
+sources = os.path.exists(PREFERENCES_PATH) and \
+          readPlist(PREFERENCES_PATH).get('Sources', []) or []
 
+def update_all():
+    for record in Paper.library_records[its.kind=='PDF']():
+        title = record.title()
+        hint, no_match = extract_date(title)
+        source = extract_source(title, hint)
+
+        contents = record.contents()
+        if re.search(re.escape(source), contents, re.IGNORECASE):
+            if source in sources:
+                sources.remove(source)
+            sources.append(source)
+
+        extracted, no_match = extract_date(contents, hint)
+
+        if not extracted:
+            print title, hint
+            for nm in no_match:
+                print '  no match', nm
+            if not hint:
+                continue
+
+        record.creation_date.set(extracted or hint)
+
+    sources.reverse() # most recently referenced ones at top
+
+def scan_one():
+    Acrobat = app(id='com.adobe.Acrobat.Pro')
+    SystemEvents = app(id='com.apple.systemevents')
+    acro_process = SystemEvents.application_processes[u'Acrobat']
+
+    filename = datetime.now().strftime('Scanned Document %y%m%d %H%m%S')
+
+    SA = ScriptingAddition()
+    SA.activate()
+    try:
+        while True:
+            result = SA.display_dialog('How many pages do you wish to scan?',
+                                       buttons=['Cancel', 'Scan'],
+                                       cancel_button=1, default_button=2,
+                                       default_answer='1')
+            try:
+                pages = int(result[k.text_returned])
+            except ValueError:
+                continue
+            if pages > 0:
+                break
+    except CommandError:
+        return False
+
+    Acrobat.activate()
+
+    acro_process.menu_bars[1].menu_bar_items['Document'].menus[1].\
+        menu_items['Scan to PDF...'].click()
+    acro_process.windows['Acrobat Scan'].buttons['Scan'].click()
+
+    # pause (Carbon -> Cocoa? use keystrokes instead?)
+    acro_process.windows['Save Scanned File As'].text_fields[1].value.\
+        set(filename)
+    acro_process.windows['Save Scanned File As'].buttons['Save'].click()
+
+    acro_scan_window = acro_process.windows['Acrobat Scan']
+
+    while True:
+        acro_process.windows['DSmobile 600'].buttons['Scan'].click()
+        while not acro_scan_window.exists():
+            time.sleep(0.1)
+
+        pages -= 1
+
+        if pages == 0:
+            acro_scan_window.groups[1].radio_buttons[2].click()
+            acro_scan_window.buttons['OK'].click()
+            break
+
+        acro_scan_window.groups[1].radio_buttons[1].click()
+        acro_scan_window.buttons['OK'].click()
+
+    scanned_document = Acrobat.documents['%s.pdf' % filename]
+    scanned_file = scanned_document.file_alias()
+    scanned_document.close()
+
+    record = Paper.import_(files=[scanned_file])[0]
     contents = record.contents()
-    extracted, no_match = extract_date(contents, hint)
+    m = re.search('(%s)' % '|'.join(map(re.escape, sources)), contents,
+                  re.IGNORECASE)
+    if m:
+        # use the saved source's case
+        title = sources[map(str.lower, sources).index(m.group(1).lower())]
+    else:
+        title = '???'
 
-    if not extracted:
-        print title, hint
-        for nm in no_match:
-            print '  no match', nm
-        if not hint:
-            continue
+    extracted, no_match = extract_date(contents)
+    if extracted:
+        title += extracted.strftime(' %Y-%m')
+        record.creation_date.set(extracted)
 
-    record.creation_date.set(extracted or hint)
+    record.title.set(title)
+
+    return True
+
+# update_all()
+
+# XXX incremental source recording from EagleFiler (use tag to record)
+
+while scan_one():
+    pass
+
+writePlist({'Sources': sources}, PREFERENCES_PATH)