Raise exceptions instead of hard exiting

11 files changed, 148 insertions, 32 deletions

diff --git a/paper2remarkable/exceptions.py b/paper2remarkable/exceptions.py
new file mode 100644
index 0000000..f287ce1
--- /dev/null
+++ b/paper2remarkable/exceptions.py
@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+
+"""Exceptions for paper2remarkable
+
+"""
+
+from . import GITHUB_URL
+
+from subprocess import CalledProcessError
+
+
+class Error(Exception):
+    """Base class for exceptions in p2r."""
+
+
+class URLResolutionError(Error):
+    """Exception raised when the abstract/pdf urls can't be resolved.
+
+    Attributes
+    ----------
+    provider : str
+        Name of the provider where the error occurred
+
+    url : str
+        Original url provided
+
+    """
+
+    def __init__(self, provider, url, reason=None):
+        self.provider = provider
+        self.url = url
+        self.reason = None
+
+    def __str__(self):
+        msg = "ERROR: Couldn't figure out {provider} URLs from provided url: {url}".format(
+            provider=self.provider, url=self.url
+        )
+        if self.reason:
+            msg += "\nReason: {reason}".format(reason=self.reason)
+        msg += "\nIf you think this might be a bug, please raise an issue on GitHub at {url}".format(
+            url=GITHUB_URL
+        )
+        return msg
+
+
+class FilenameMissingError(Error):
+    """Exception raised for providers that need a filename to be provided"""
+
+    def __init__(self, provider):
+        self.provider = provider
+
+    def __str__(self):
+        msg = "ERROR: Filename must be given with the {provider} provider (hint: use --filename)".format(
+            provider=self.provider
+        )
+        msg += "\nIf you think this might be a bug, please raise an issue on GitHub at {url}".format(
+            url=GITHUB_URL
+        )
+        return msg
+
+
+class FileTypeError(Error):
+    """Exception raised when we have a mismatch in filetype."""
+
+    def __init__(self, filename, filetype):
+        self.filename = filename
+        self.filetype = filetype
+
+    def __str__(self):
+        msg = "ERROR: File {filename} isn't of type {filetype}.".format(
+            filename=self.filename, filetype=self.filetype
+        )
+        msg += "\nIf you think this might be a bug, please raise an issue on GitHub at {url}".format(
+            url=GITHUB_URL
+        )
+        return msg
+
+
+class RemarkableError(Error):
+    """Exceptions raised when interacting with the reMarkable fails."""
+
+    def __init__(self, message):
+        self.message = message
+
+    def __str__(self):
+        msg = "ERROR: {message}".format(message=self.message)
+        msg += "\nIf you think this might be a bug, please raise an issue on GitHub at {url}".format(
+            url=GITHUB_URL
+        )
+        return msg
+
+
+class _CalledProcessError(CalledProcessError):
+    """Exception raised when subprocesses fail.
+
+    We subclass the CalledProcessError so we can add our custom error message.
+    """
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+
+    def __str__(self):
+        parent = super().__str__()
+        msg = (
+            parent
+            + "\nIf you think this might be a bug, please raise an issue on GitHub at {url}".format(
+                url=GITHUB_URL
+            )
+        )
+        return msg
diff --git a/paper2remarkable/providers/acm.py b/paper2remarkable/providers/acm.py
index 08e8cb2..45eeacf 100644
--- a/paper2remarkable/providers/acm.py
+++ b/paper2remarkable/providers/acm.py
@@ -12,9 +12,9 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from .. import GITHUB_URL
-from ..utils import exception
+from ..exceptions import URLResolutionError
 from ..log import Logger
+from ..utils import exception
 
 logger = Logger()
 
@@ -57,7 +57,7 @@ class ACM(Provider):
         m = re.match(self.re_abs, url) or re.match(self.re_pdf, url)
         if m:
             return m["doi"]
-        exception("Couldn't retrieve ACM publication DOI.")
+        raise URLResolutionError("ACM", url, reason="Failed to retrieve DOI.")
 
     def get_abs_pdf_urls(self, url):
         if re.match(self.re_abs, url):
@@ -71,9 +71,7 @@ class ACM(Provider):
             doi = self._get_doi(url)
             abs_url = "https://dl.acm.org/doi/{doi}".format(doi=doi)
         else:
-            exception(
-                "Couldn't figure out ACM urls from provided url: %s" % url
-            )
+            raise URLResolutionError("ACM", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/arxiv.py b/paper2remarkable/providers/arxiv.py
index 1fd1795..282eb09 100644
--- a/paper2remarkable/providers/arxiv.py
+++ b/paper2remarkable/providers/arxiv.py
@@ -14,7 +14,10 @@ import subprocess
 
 from ._info import Informer
 from ._base import Provider
-from ..utils import exception
+from ..exceptions import (
+    URLResolutionError,
+    _CalledProcessError as CalledProcessError,
+)
 from ..log import Logger
 
 logger = Logger()
@@ -49,7 +52,7 @@ class Arxiv(Provider):
             abs_url = url[:-4].replace("pdf", "abs")
             pdf_url = url
         else:
-            exception("Couldn't figure out arXiv urls.")
+            raise URLResolutionError("arXiv", url)
         return abs_url, pdf_url
 
     def validate(src):
@@ -72,7 +75,9 @@ class Arxiv(Provider):
             ]
         )
         if not status == 0:
-            exception("pdftk failed to uncompress the pdf.")
+            raise CalledProcessError(
+                "pdftk failed to uncompress the PDF file."
+            )
 
         with open(uncompress_file, "rb") as fid:
             data = fid.read()
@@ -94,6 +99,6 @@ class Arxiv(Provider):
             [self.pdftk_path, removed_file, "output", output_file, "compress"]
         )
         if not status == 0:
-            exception("pdftk failed to compress the pdf.")
+            raise CalledProcessError("pdftk failed to compress the PDF file.")
 
         return output_file
diff --git a/paper2remarkable/providers/citeseerx.py b/paper2remarkable/providers/citeseerx.py
index fdc0e8a..82adca7 100644
--- a/paper2remarkable/providers/citeseerx.py
+++ b/paper2remarkable/providers/citeseerx.py
@@ -12,7 +12,7 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class CiteSeerXInformer(Informer):
@@ -38,7 +38,9 @@ class CiteSeerX(Provider):
         m = re.match(self.re_abs, url) or re.match(self.re_pdf, url)
         if m:
             return m["doi"]
-        exception("Couldn't retrieve CiteSeerX publication doi.")
+        raise URLResolutionError(
+            "CiteSeerX", url, reason="Failed to retrieve DOI."
+        )
 
     def get_abs_pdf_urls(self, url):
         """ Get the pdf and abstract url from a OpenReview url """
@@ -55,7 +57,7 @@ class CiteSeerX(Provider):
                 doi=doi
             )
         else:
-            exception("Couldn't figure out CiteSeerX urls.")
+            raise URLResolutionError("CiteSeerX", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/neurips.py b/paper2remarkable/providers/neurips.py
index 9dce4a5..87cf2c1 100644
--- a/paper2remarkable/providers/neurips.py
+++ b/paper2remarkable/providers/neurips.py
@@ -12,7 +12,7 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class NeurIPSInformer(Informer):
@@ -41,7 +41,7 @@ class NeurIPS(Provider):
             abs_url = url.replace(".pdf", "")
             pdf_url = url
         else:
-            exception("Couldn't figure out NeurIPS urls.")
+            raise URLResolutionError("NeurIPS", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/openreview.py b/paper2remarkable/providers/openreview.py
index bfb139d..47c0555 100644
--- a/paper2remarkable/providers/openreview.py
+++ b/paper2remarkable/providers/openreview.py
@@ -12,7 +12,7 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class OpenReviewInformer(Informer):
@@ -41,7 +41,7 @@ class OpenReview(Provider):
             abs_url = url.replace("pdf", "forum")
             pdf_url = url
         else:
-            exception("Couldn't figure out OpenReview urls.")
+            raise URLResolutionError("OpenReview", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/pdf_url.py b/paper2remarkable/providers/pdf_url.py
index d80b1a9..5314ec7 100644
--- a/paper2remarkable/providers/pdf_url.py
+++ b/paper2remarkable/providers/pdf_url.py
@@ -12,15 +12,13 @@ import urllib
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import FilenameMissingError
 
 
 class PdfUrlInformer(Informer):
     def get_filename(self, abs_url):
-        # if this is called, filename must not be provided
-        exception(
-            "Filename must be provided with PDFUrlProvider (use --filename)"
-        )
+        # if this is called, filename must not have been provided
+        raise FilenameMissingError(provider="PDFUrl")
 
 
 class PdfUrl(Provider):
diff --git a/paper2remarkable/providers/pmlr.py b/paper2remarkable/providers/pmlr.py
index 6981c3b..7b76d8a 100644
--- a/paper2remarkable/providers/pmlr.py
+++ b/paper2remarkable/providers/pmlr.py
@@ -12,7 +12,7 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class PMLRInformer(Informer):
@@ -20,7 +20,7 @@ class PMLRInformer(Informer):
     meta_date_key = "citation_publication_date"
 
     def _format_authors(self, soup_authors):
-        return super()._format_authors(soup_authors, sep=' ', idx=-1)
+        return super()._format_authors(soup_authors, sep=" ", idx=-1)
 
 
 class PMLR(Provider):
@@ -56,7 +56,7 @@ class PMLR(Provider):
             abs_url = "/".join(parts[:-1]) + ".html"
             pdf_url = url
         else:
-            exception("Couldn't figure out PLMR urls.")
+            raise URLResolutionError("PMLR", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/pubmed.py b/paper2remarkable/providers/pubmed.py
index ba4cca0..5387cb0 100644
--- a/paper2remarkable/providers/pubmed.py
+++ b/paper2remarkable/providers/pubmed.py
@@ -12,7 +12,7 @@ import re
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class PubMedInformer(Informer):
@@ -50,7 +50,7 @@ class PubMed(Provider):
             abs_url = url
             pdf_url = url.rstrip("/") + "/pdf"  # it redirects, usually
         else:
-            exception("Couldn't figure out PMC urls.")
+            raise URLResolutionError("PMC", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/providers/springer.py b/paper2remarkable/providers/springer.py
index ce4acdd..5ce2564 100644
--- a/paper2remarkable/providers/springer.py
+++ b/paper2remarkable/providers/springer.py
@@ -13,7 +13,7 @@ import urllib
 
 from ._base import Provider
 from ._info import Informer
-from ..utils import exception
+from ..exceptions import URLResolutionError
 
 
 class SpringerInformer(Informer):
@@ -42,7 +42,7 @@ class Springer(Provider):
             abs_url = url.replace("content/pdf", "article")[: -len(".pdf")]
             pdf_url = urllib.parse.unquote(url)
         else:
-            exception("Couldn't figure out Springer urls.")
+            raise URLResolutionError("Springer", url)
         return abs_url, pdf_url
 
     def validate(src):
diff --git a/paper2remarkable/utils.py b/paper2remarkable/utils.py
index de27973..7d68964 100644
--- a/paper2remarkable/utils.py
+++ b/paper2remarkable/utils.py
@@ -18,6 +18,7 @@ import unidecode
 
 from . import GITHUB_URL
 from .log import Logger
+from .exceptions import FileTypeError, RemarkableError
 
 HEADERS = {
     "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) "
@@ -64,7 +65,7 @@ def assert_file_is_pdf(filename):
         del pdf
         return True
     except PyPDF2.utils.PdfReadError:
-        exception("File %s isn't a valid pdf file." % filename)
+        raise FileTypeError(filename, "pdf")
 
 
 def download_url(url, filename):
@@ -122,7 +123,7 @@ def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
             stdout=subprocess.DEVNULL,
         )
         if not status == 0:
-            exception(
+            raise RemarkableError(
                 "Creating directory %s on reMarkable failed" % remarkable_dir
             )
 
@@ -132,5 +133,7 @@ def upload_to_remarkable(filepath, remarkable_dir="/", rmapi_path="rmapi"):
         stdout=subprocess.DEVNULL,
     )
     if not status == 0:
-        exception("Uploading file %s to reMarkable failed" % filepath)
+        raise RemarkableError(
+            "Uploading file %s to reMarkable failed" % filepath
+        )
     logger.info("Upload successful.")