Option to add site wide search

[sahiljhawar]

To implement a site wide search instead of just the options shown in the search menu, like searching for in-text content? @george-gca Says it will be slower which I can agree with (however I am not familiar with JS and webdev).

If this feature is already there please let me know or if this is something which needs to built from scratch?

[george-gca]

This needs to be "built from scratch" but it is not that hard. Two changes are needed:

1. Add in _includes/scripts/search.liquid inside the elements that you want the content to be searched (e.g.: posts) a new field with the contents, something like contents: "{{ post.content | strip_html | strip_newlines | escape }}",
2. in assets/js/search/ninja-keys.min.js search for this string: const s=commandScore(e.title+" "+e.description,n); and add e.contents to the commandScore: const s=commandScore(e.title+" "+e.description+" "+e.contents,n);

[george-gca]

Actually this changes only the results computed, not the text displayed in the results. What this means is, if you search for some text that is inside a post, the referred post should be placed first in the search, but the text that contains the search will not appear in the search result itself. This requires more changes in the component, and I don't know where those changes should be made.

[sahiljhawar]

I am fine with this funtionality as well. But the problem is the search does not return anything.

[george-gca]

Take a look at the browser console when looking at your page. Probably you are having trouble with some of the pages' contents. For example, if inside the page you have a double quotes, it will conflict with the double quotes used in the javascript, creating something like:

contents: "Just wrap your images with <div class="col-sm"",

To avoid this you should do some cleaning on the contents, like:

contents: "{{ item.content | strip_html | strip_newlines | replace: '"', "'" | escape | strip }}",

[sahiljhawar]

contents: "Just wrap your images with <div class="col-sm"",

No, there are no errors in the browser console

contents: "{{ item.content | strip_html | strip_newlines | replace: '"', "'" | escape | strip }}",

after adding this, again there are no changes and the following attribute error occurs in terminal

Console error

Traceback (most recent call last):
  File "/Users/jhawarji/opt/anaconda3/bin/jupyter-nbconvert", line 11, in <module>
    sys.exit(main())
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/jupyter_core/application.py", line 269, in launch_instance
    return super().launch_instance(argv=argv, **kwargs)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/traitlets/config/application.py", line 846, in launch_instance
    app.start()
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/nbconvertapp.py", line 369, in start
    self.convert_notebooks()
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/nbconvertapp.py", line 541, in convert_notebooks
    self.convert_single_notebook(notebook_filename)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/nbconvertapp.py", line 506, in convert_single_notebook
    output, resources = self.export_single_notebook(notebook_filename, resources, input_buffer=input_buffer)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/nbconvertapp.py", line 435, in export_single_notebook
    output, resources = self.exporter.from_filename(notebook_filename, resources=resources)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/exporters/exporter.py", line 190, in from_filename
    return self.from_file(f, resources=resources, **kw)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/exporters/exporter.py", line 208, in from_file
    return self.from_notebook_node(nbformat.read(file_stream, as_version=4), resources=resources, **kw)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/exporters/html.py", line 210, in from_notebook_node
    return super().from_notebook_node(nb, resources, **kw)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/exporters/templateexporter.py", line 392, in from_notebook_node
    output = self.template.render(nb=nb_copy, resources=resources)
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/jinja2/environment.py", line 1301, in render
    self.environment.handle_exception()
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/jinja2/environment.py", line 936, in handle_exception
    raise rewrite_traceback_stack(source=source)
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/lab/index.html.j2", line 3, in top-level template code
    {% from 'jupyter_widgets.html.j2' import jupyter_widgets %}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/lab/base.html.j2", line 260, in top-level template code
    {% set div_id = uuid4() %}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/base/display_priority.j2", line 1, in top-level template code
    {%- extends 'base/null.j2' -%}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/base/null.j2", line 24, in top-level template code
    {%- block header -%}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/lab/index.html.j2", line 9, in block 'header'
    {%- block html_head -%}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/lab/index.html.j2", line 36, in block 'html_head'
    {% block notebook_css %}
  File "/Users/jhawarji/opt/anaconda3/share/jupyter/nbconvert/templates/lab/index.html.j2", line 37, in block 'notebook_css'
    {{ resources.include_css("static/index.css") }}
  File "/Users/jhawarji/opt/anaconda3/lib/python3.9/site-packages/nbconvert/exporters/html.py", line 216, in resources_include_css
    return jinja2.Markup(code)
AttributeError: module 'jinja2' has no attribute 'Markup'

[george-gca]

The problem is probably due to a post with a jupyter notebook inside. These edge cases will be hard to handle. This is another reason why I decided to just include a search within titles and descriptions. Also the search works more like another form of navigation in the site, since it can all be done without a mouse.