Add eol --newline option for text content in nbconvert.writers.files

CHANGELOG.md

@@ -1586,6 +1586,7 @@ raw template
 {%- endblock in_prompt -%}
     """
 
+
 exporter_attr = AttrExporter()
 output_attr, _ = exporter_attr.from_notebook_node(nb)
 assert "raw template" in output_attr

nbconvert/nbconvertapp.py

@@ -17,7 +17,16 @@
 from textwrap import dedent, fill
 
 from jupyter_core.application import JupyterApp, base_aliases, base_flags
-from traitlets import Bool, DottedObjectName, Instance, List, Type, Unicode, default, observe
+from traitlets import (
+    Bool,
+    DottedObjectName,
+    Instance,
+    List,
+    Type,
+    Unicode,
+    default,
+    observe,
+)
 from traitlets.config import Configurable, catch_config_error
 from traitlets.utils.importstring import import_item
 
@@ -60,6 +69,7 @@ def validate(self, obj, value):
         "writer": "NbConvertApp.writer_class",
         "post": "NbConvertApp.postprocessor_class",
         "output": "NbConvertApp.output_base",
+        "newline": "NbConvertApp.newline",
         "output-dir": "FilesWriter.build_directory",
         "reveal-prefix": "SlidesExporter.reveal_url_prefix",
         "nbformat": "NotebookExporter.nbformat_version",
@@ -120,7 +130,10 @@ def validate(self, obj, value):
         ),
         "coalesce-streams": (
             {
-                "NbConvertApp": {"use_output_suffix": False, "export_format": "notebook"},
+                "NbConvertApp": {
+                    "use_output_suffix": False,
+                    "export_format": "notebook",
+                },
                 "FilesWriter": {"build_directory": ""},
                 "CoalesceStreamsPreprocessor": {"enabled": True},
             },
@@ -186,6 +199,11 @@ def validate(self, obj, value):
             },
             """Whether the HTML in Markdown cells and cell outputs should be sanitized..""",
         ),
+        "use-crlf-newline": (
+            {"NbConvertApp": {"newline": "\r\n"}},
+            "Implies CLRF newline.",
+        ),
+        "use-lf-newline": ({"NbConvertApp": {"newline": "\n"}}, "Implies LF newline."),
     }
 )
 
@@ -311,6 +329,12 @@ def _writer_class_changed(self, change):
             new = self.writer_aliases[new.lower()]
         self.writer_factory = import_item(new)
 
+    newline = Unicode(
+        None,
+        help="""The line ending to use when writing text, defaults to builtin:`open`""",
+        allow_none=True,
+    ).tag(config=True)
+
     # Post-processor specific variables
     postprocessor = Instance(
         "nbconvert.postprocessors.base.PostProcessorBase",
@@ -523,7 +547,9 @@ def write_single_notebook(self, output, resources):
         if not self.writer:
             msg = "No writer object defined!"
             raise ValueError(msg)
-        return self.writer.write(output, resources, notebook_name=notebook_name)
+        return self.writer.write(
+            output, resources, notebook_name=notebook_name, newline=self.newline
+        )
 
     def postprocess_single_notebook(self, write_results):
         """Step 4: Post-process the written file
@@ -628,7 +654,13 @@ def document_config_options(self):
         """
         categories = {
             category: [c for c in self._classes_inc_parents() if category in c.__name__.lower()]
-            for category in ["app", "exporter", "writer", "preprocessor", "postprocessor"]
+            for category in [
+                "app",
+                "exporter",
+                "writer",
+                "preprocessor",
+                "postprocessor",
+            ]
         }
         accounted_for = {c for category in categories.values() for c in category}
         categories["other"] = [c for c in self._classes_inc_parents() if c not in accounted_for]

nbconvert/writers/files.py

@@ -77,7 +77,7 @@ def _write_items(self, items, build_dir):
             with open(dest, "wb") as f:
                 f.write(data)
 
-    def write(self, output, resources, notebook_name=None, **kw):
+    def write(self, output, resources, notebook_name=None, newline=None, **kw):
         """
         Consume and write Jinja output to the file system.  Output directory
         is set via the 'build_directory' variable of this instance (a
@@ -147,11 +147,25 @@ def write(self, output, resources, notebook_name=None, **kw):
         dest_path = Path(build_directory) / dest
 
         # Write conversion results.
-        self.log.info("Writing %i bytes to %s", len(output), dest_path)
+        if "\\\\" in repr(newline):
+            self.log.warning(
+                "Argument 'newline' contains escaped escape characters: %s",
+                format(repr(newline)),
+            )
+
         if isinstance(output, str):
-            with open(dest_path, "w", encoding="utf-8") as f:
+            self.log.info(
+                "Writing %i len string to %s, (repr(newline): %s)",
+                len(output),
+                dest_path,
+                repr(newline),
+            )
+            with open(dest_path, "w", encoding="utf-8", newline=newline) as f:
                 f.write(output)
+
         else:
+            # newline only applies in text mode.
+            self.log.info("Writing %i bytes to %s", len(output), dest_path)
             with open(dest_path, "wb") as f:
                 f.write(output)