Looking for text.pdf that does not exist

[tadamhicks]

When I set up a watch in a directory, I frequently see

ERROR: Cannot find specified pdf file /path/to/file_text.pdf

This does not happen with every pdf I place in there, but when it does the watchdog tips over.

I don't know what it's looking for, and I would like for the watchdog to stay running despite errors.

[palica]

You can patch pypdfocr_watcher.py like this:

diff --git a/pypdfocr/pypdfocr_watcher.py b/pypdfocr/pypdfocr_watcher.py
index 73581e0..cb78002 100755
--- a/pypdfocr/pypdfocr_watcher.py
+++ b/pypdfocr/pypdfocr_watcher.py
@@ -93,7 +93,7 @@ class PyPdfWatcher(FileSystemEventHandler):

         """
         if ev_path.endswith(".pdf"):
-            if not ev_path.endswith("_ocr.pdf"):
+            if not ev_path.endswith(("_ocr.pdf","_text.pdf")):^M
                 PyPdfWatcher.events_lock.acquire()
                 if not ev_path in PyPdfWatcher.events:
                     PyPdfWatcher.events[ev_path] = time.time()

[virantha]

Thanks, I'll try to get this into the next release.

[mmatiaschek]

I get this error regularly - a release of 0.9.1 would be great!

[mmatiaschek]

Patch works fine, i incorporated it into my docker image...
https://hub.docker.com/r/mmatiaschek/pypdfocr/

[danmash]

@mmatiaschek thank you!

[danmash]

@virantha You made a typo in pypdfocr_watcher.py
if not ev_path.endswith(("_ocr.pdf", "_test.pdf")):

_test instead of _text

[mattobrien415]

I am still having this problem using 0.9.1! My monitoring dies. I checked pypdfocr_watcher.py, the fix is clearly there. Am I the only one?

Only thing I can think of, is that I'm running the process over multiple directories at once.

[solartune]

I have the same problem. @virantha please, check the line
if not ev_path.endswith(("_ocr.pdf", "_test.pdf")):
There should be _text as said @danmash