Hotfixing idefics base64 parsing. (#1103)

# What does this PR do?

<!--
Congratulations! You've made it this far! You're not quite done yet
though.

Once merged, your PR is going to appear in the release notes with the
title you set, so make sure it's a great title that fully reflects the
extent of your awesome contribution.

Then, please replace this with a description of the change and which
issue is fixed (if applicable). Please also include relevant motivation
and context. List any dependencies (if any) that are required for this
change.

Once you're done, someone will review your PR shortly (see the section
"Who can review?" below to tag some potential reviewers). They may
suggest changes to make the code even better. If no one reviewed your PR
after a week has passed, don't hesitate to post a new comment
@-mentioning the same persons---sometimes notifications get lost.
-->

<!-- Remove if not applicable -->

Fixes # (issue)


## Before submitting
- [ ] This PR fixes a typ...

server/text_generation_server/models/custom_modeling/idefics_image_processing.py

@@ -201,8 +201,13 @@ class IdeficsImageProcessor(BaseImageProcessor):
                 response = requests.get(image_url_or_urls, stream=True, headers=headers, timeout=(1, 5))
                 response.raise_for_status()
                 content = response.content
-            else:
+            elif image.startswith("data:"):
+                # https://stackoverflow.com/questions/17090571/is-there-a-way-to-set-background-image-as-a-base64-encoded-image
+                # data:image/png;base64,xxx
+                image = image.split(",")[-1]
                 content = base64.b64decode(image)
+            else:
+                raise ValueError(f"Unrecognized image {image}")
 
             try:
                 image = Image.open(BytesIO(content))

server/text_generation_server/models/custom_modeling/idefics_processing.py

@@ -112,6 +112,11 @@ def is_url(string):
     result = urlparse(string)
     return all([result.scheme, result.netloc])
 
+def is_image(string):
+    """Checks if the passed string contains a valid url and nothing else. e.g. if space is included it's immediately
+    invalidated the url"""
+    return is_url(string) or string.startswith("data:")
+
 
 class IdeficsProcessor(ProcessorMixin):
     r"""
@@ -314,7 +319,7 @@ class IdeficsProcessor(ProcessorMixin):
 
                 if isinstance(item, str):
                     item = item.strip(" ")
-                    if is_url(item):
+                    if is_image(item):
                         image = self.image_processor.fetch_images(item)
                         full_text += image_tokens(last_was_image)
                         image_objects.append(image)
@@ -339,6 +344,7 @@ class IdeficsProcessor(ProcessorMixin):
 
             image_objects = self.image_processor(image_objects, transform=transform)
 
+
             text_encoding = self.tokenizer(
                 text=full_text,
                 add_special_tokens=False,