Switch to a simple logger singleton

The logging module gave problems because one
of the pdf packages is based on a package
that extensively used the info level of the
logging module, and this seemed like the
easiest solution.

8 files changed, 95 insertions, 28 deletions

diff --git a/paper2remarkable/log.py b/paper2remarkable/log.py
new file mode 100644
index 0000000..bae1cbf
--- /dev/null
+++ b/paper2remarkable/log.py
@@ -0,0 +1,56 @@
+# -*- coding: utf-8 -*-
+
+"""Just a simple logger
+
+Author: G.J.J. van den Burg
+License: See LICENSE file.
+Copyright: 2019, G.J.J. van den Burg
+
+"""
+
+# NOTE: I know about the logging module, but this was easier because one of the
+# dependencies was using that and it became complicated. This one is obviously
+# not thread-safe and is very simple.
+
+import datetime
+import sys
+
+
+class Singleton(type):
+    # https://stackoverflow.com/q/6760685
+    _instances = {}
+
+    def __call__(cls, *args, **kwargs):
+        if cls not in cls._instances:
+            cls._instances[cls] = super(Singleton, cls).__call__(
+                *args, **kwargs
+            )
+        return cls._instances[cls]
+
+
+class Logger(metaclass=Singleton):
+    def __init__(self):
+        self.enabled = True
+
+    def enable(self):
+        self.enabled = True
+
+    def disable(self):
+        self.enabled = False
+
+    def _log(self, msg, mode):
+        if not self.enabled:
+            return
+        if not mode in ("info", "warn"):
+            raise ValueError("Unknown logging mode: %s" % mode)
+        file = sys.stdout if mode == "info" else sys.stderr
+        now = datetime.datetime.now()
+        nowstr = now.strftime("%Y-%m-%d %H:%M:%S")
+        print("%s - %s - %s" % (nowstr, mode.upper(), msg), file=file)
+        file.flush()
+
+    def info(self, msg):
+        self._log(msg, "info")
+
+    def warning(self, msg):
+        self._log(msg, "warn")
diff --git a/paper2remarkable/pdf_ops.py b/paper2remarkable/pdf_ops.py
index d1eae40..8636017 100644
--- a/paper2remarkable/pdf_ops.py
+++ b/paper2remarkable/pdf_ops.py
@@ -10,27 +10,28 @@ Copyright: 2019, The Alan Turing Institute
 
 
 import PyPDF2
-import logging
 import os
 import subprocess
 
 from .crop import Cropper
+from .log import Logger
 
+logger = Logger()
 
 def crop_pdf(filepath, pdfcrop_path="pdfcrop"):
     """Crop the pdf file using Cropper
     """
-    logging.info("Cropping pdf file")
+    logger.info("Cropping pdf file")
     cropped_file = os.path.splitext(filepath)[0] + "-crop.pdf"
 
     cropper = Cropper(filepath, cropped_file, pdfcrop_path=pdfcrop_path)
     status = cropper.crop(margins=15)
 
     if not status == 0:
-        logging.warning("Failed to crop the pdf file at: %s" % filepath)
+        logger.warning("Failed to crop the pdf file at: %s" % filepath)
         return filepath
     if not os.path.exists(cropped_file):
-        logging.warning(
+        logger.warning(
             "Can't find cropped file '%s' where expected." % cropped_file
         )
         return filepath
@@ -40,17 +41,17 @@ def crop_pdf(filepath, pdfcrop_path="pdfcrop"):
 def center_pdf(filepath, pdfcrop_path="pdfcrop"):
     """Center the pdf file on the reMarkable
     """
-    logging.info("Centering pdf file")
+    logger.info("Centering pdf file")
     centered_file = os.path.splitext(filepath)[0] + "-center.pdf"
 
     cropper = Cropper(filepath, centered_file, pdfcrop_path=pdfcrop_path)
     status = cropper.center()
 
     if not status == 0:
-        logging.warning("Failed to center the pdf file at: %s" % filepath)
+        logger.warning("Failed to center the pdf file at: %s" % filepath)
         return filepath
     if not os.path.exists(centered_file):
-        logging.warning(
+        logger.warning(
             "Can't find centered file '%s' where expected." % centered_file
         )
         return filepath
@@ -60,7 +61,7 @@ def center_pdf(filepath, pdfcrop_path="pdfcrop"):
 def blank_pdf(filepath):
     """Add blank pages to PDF
     """
-    logging.info("Adding blank pages")
+    logger.info("Adding blank pages")
     input_pdf = PyPDF2.PdfFileReader(filepath)
     output_pdf = PyPDF2.PdfFileWriter()
     for page in input_pdf.pages:
@@ -76,7 +77,7 @@ def blank_pdf(filepath):
 def shrink_pdf(filepath, gs_path="gs"):
     """Shrink the PDF file size using Ghostscript
     """
-    logging.info("Shrinking pdf file")
+    logger.info("Shrinking pdf file")
     output_file = os.path.splitext(filepath)[0] + "-shrink.pdf"
     status = subprocess.call(
         [
@@ -92,6 +93,6 @@ def shrink_pdf(filepath, gs_path="gs"):
         ]
     )
     if not status == 0:
-        logging.warning("Failed to shrink the pdf file")
+        logger.warning("Failed to shrink the pdf file")
         return filepath
     return output_file
diff --git a/paper2remarkable/providers/_base.py b/paper2remarkable/providers/_base.py
index ca6ab70..5432d48 100644
--- a/paper2remarkable/providers/_base.py
+++ b/paper2remarkable/providers/_base.py
@@ -9,7 +9,6 @@ Copyright: 2019, G.J.J. van den Burg
 """
 
 import abc
-import logging
 import os
 import shutil
 import tempfile
@@ -17,6 +16,9 @@ import tempfile
 from ._info import Informer
 from ..pdf_ops import crop_pdf, center_pdf, blank_pdf, shrink_pdf
 from ..utils import assert_file_is_pdf, download_url, upload_to_remarkable
+from ..log import Logger
+
+logger = Logger()
 
 
 class Provider(metaclass=abc.ABCMeta):
@@ -45,9 +47,8 @@ class Provider(metaclass=abc.ABCMeta):
         self.informer = Informer()
 
         # disable logging if requested
-        logging.basicConfig(level=logging.INFO)
         if not verbose:
-            logging.disable()
+            logger.disable()
 
         # Define the operations to run on the pdf. Providers can add others.
         self.operations = [("crop", self.crop_pdf)]
@@ -58,7 +59,7 @@ class Provider(metaclass=abc.ABCMeta):
             self.operations.append(("blank", blank_pdf))
         self.operations.append(("shrink", self.shrink_pdf))
 
-        logging.info("Starting %s" % type(self).__name__)
+        logger.info("Starting %s" % type(self).__name__)
 
     @staticmethod
     @abc.abstractmethod
diff --git a/paper2remarkable/providers/_info.py b/paper2remarkable/providers/_info.py
index 04efcb1..9130e34 100644
--- a/paper2remarkable/providers/_info.py
+++ b/paper2remarkable/providers/_info.py
@@ -3,12 +3,14 @@
 """Functionality for retrieving paper info
 """
 
-import logging
 import titlecase
 import unidecode
 import bs4
 
 from ..utils import clean_string, get_page_with_retry
+from ..log import Logger
+
+logger = Logger()
 
 
 class Informer:
@@ -38,7 +40,7 @@ class Informer:
         The provided url must be to a HTMl page where this information can be 
         found, not to the PDF file itself.
         """
-        logging.info("Generating output filename")
+        logger.info("Generating output filename")
 
         # Retrieve the paper information
         self.get_info(abs_url)
@@ -59,11 +61,11 @@ class Informer:
 
         name = authors + "_-_" + title + "_" + year + ".pdf"
         name = unidecode.unidecode(name)
-        logging.info("Created filename: %s" % name)
+        logger.info("Created filename: %s" % name)
         return name
 
     def get_info(self, url):
-        logging.info("Getting paper info")
+        logger.info("Getting paper info")
         page = get_page_with_retry(url)
         soup = bs4.BeautifulSoup(page, "html.parser")
         self.authors = self.authors or self.get_authors(soup)
diff --git a/paper2remarkable/providers/acm.py b/paper2remarkable/providers/acm.py
index e14efa7..a0d79bd 100644
--- a/paper2remarkable/providers/acm.py
+++ b/paper2remarkable/providers/acm.py
@@ -15,6 +15,9 @@ from ._base import Provider
 from ._info import Informer
 from .. import GITHUB_URL
 from ..utils import exception, get_page_with_retry
+from ..log import Logger
+
+logger = Logger()
 
 
 class ACMInformer(Informer):
@@ -26,7 +29,7 @@ class ACMInformer(Informer):
 
     def _format_year(self, soup_date):
         if not re.match("\d{2}/\d{2}/\d{4}", soup_date.strip()):
-            self.warn(
+            logger.warning(
                 "Couldn't extract year from ACM page, please raise an "
                 "issue on GitHub so it can be fixed: %s" % GITHUB_URL
             )
diff --git a/paper2remarkable/providers/arxiv.py b/paper2remarkable/providers/arxiv.py
index d950e47..e022658 100644
--- a/paper2remarkable/providers/arxiv.py
+++ b/paper2remarkable/providers/arxiv.py
@@ -11,11 +11,13 @@ Copyright: 2019, G.J.J. van den Burg
 import os
 import re
 import subprocess
-import logging
 
 from ._info import Informer
 from ._base import Provider
 from ..utils import exception
+from ..log import Logger
+
+logger = Logger()
 
 
 class ArxivInformer(Informer):
@@ -52,7 +54,7 @@ class Arxiv(Provider):
 
     def dearxiv(self, input_file):
         """Remove the arXiv timestamp from a pdf"""
-        logging.info("Removing arXiv timestamp")
+        logger.info("Removing arXiv timestamp")
         basename = os.path.splitext(input_file)[0]
         uncompress_file = basename + "_uncompress.pdf"
 
diff --git a/paper2remarkable/providers/pdf_url.py b/paper2remarkable/providers/pdf_url.py
index f28c742..dfc8646 100644
--- a/paper2remarkable/providers/pdf_url.py
+++ b/paper2remarkable/providers/pdf_url.py
@@ -12,7 +12,6 @@ import urllib
 
 from ._base import Provider
 from ._info import Informer
-
 from ..utils import exception
 
 class PdfUrlInformer(Informer):
diff --git a/paper2remarkable/utils.py b/paper2remarkable/utils.py
index 2bed231..d80c954 100644
--- a/paper2remarkable/utils.py
+++ b/paper2remarkable/utils.py
@@ -9,7 +9,6 @@ Copyright: 2019, G.J.J. van den Burg
 """
 
 import PyPDF2
-import logging
 import requests
 import string
 import subprocess
@@ -18,6 +17,7 @@ import time
 import unidecode
 
 from . import GITHUB_URL
+from .log import Logger
 
 HEADERS = {
     "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) "
@@ -26,6 +26,8 @@ HEADERS = {
 }
 
 
+logger = Logger()
+
 def exception(msg):
     print("ERROR: " + msg, file=sys.stderr)
     print("Error occurred. Exiting.", file=sys.stderr)
@@ -37,6 +39,7 @@ def exception(msg):
     raise SystemExit(1)
 
 
+
 def clean_string(s):
     """ Clean a string by replacing accented characters with equivalents and 
     keeping only the allowed characters (ascii letters, digits, underscore, 
@@ -64,7 +67,7 @@ def assert_file_is_pdf(filename):
 
 def download_url(url, filename):
     """Download the content of an url and save it to a filename """
-    logging.info("Downloading file at url: %s" % url)
+    logger.info("Downloading file at url: %s" % url)
     content = get_page_with_retry(url)
     with open(filename, "wb") as fid:
         fid.write(content)
@@ -80,18 +83,18 @@ def get_page_with_retry(url, tries=5):
         except requests.exceptions.ConnectionError:
             error = True
         if error or not res.ok:
-            logging.warning(
+            logger.warning(
                 "(%i/%i) Error getting url %s. Retrying in 5 seconds."
                 % (count, tries, url)
             )
             time.sleep(5)
             continue
-        logging.info("Downloading url: %s" % url)
+        logger.info("Downloading url: %s" % url)
         return res.content
 
 
 def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
-    logging.info("Starting upload to reMarkable")
+    logger.info("Starting upload to reMarkable")
 
     # Create the reMarkable dir if it doesn't exist
     remarkable_dir = remarkable_dir.rstrip("/")
@@ -112,4 +115,4 @@ def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
     )
     if not status == 0:
         exception("Uploading file %s to reMarkable failed" % filepath)
-    logging.info("Upload successful.")
+    logger.info("Upload successful.")