mukta-strot / marathi-shabd Goto Github PK

• parse the database db.csv file tags column and extract the individual tags (topics) from it.
• if there are multiple tags (separated by semicolons) those also need to be separately extracted
• duplicates should be removed
• the output of this parser should be a list of tags in a string array separated by a delimiter (comma for now)
• this string array can be later used in the main.py script to generated output files by topic (which are currently hardcoded.)

words in the database db.csv which have multiple tags that are separated by semicolon (e.g. work;engineering), do not appear in either of the markdown files (i.e. word.md and engineering.md) in the browse/topics folder.

But words that have only single tag do appear in the same files.

• input file will be the db.csv
• output will be a markdown file which will be used on the github pages website. for now it will the be the home page of the site.
• for now, a user will have to manual search for a word of interest (or can also use the browser's search function.)

Each word has one or more tags, which are like categories under which the word falls.

On the "search" page, Give user the option of searching such categories. Since the categories will be limited by those in the database, only those should be made available to search. The search (more like selection) can be done via a drop down menu.

(The list for the drop down menu itself can be created by parsing the tags present in the database.)

Once user selects a particular tag from the menu, they should be shown all the word-blocks for the words that fall under that tag.

a cache folder gets created everytime a python script is run. Is there a way to prevent this from being created? If not then can be deleted after the scripts are executed? If that too is not possible, then probably this can be added to gitignore.

• currently the topic files being created have only the word blocks for the words under that topic.
• we need to add a title/heading to each topic file, so that it is easy to understand which topic this file is about.
• this will also be helpful when combining (concatenating) multiple topic files to each other, whenever such a feature is needed in the future.
• the title should be same as name of the file

eg. the title for topic file science.md will look like

science

by using

# science

... followed by the individual word blocks

This script is needed to create a single markdown file containing all the words from the database csv file.

The output format for each word and it's Marathi equivalent is given in the example present in the template folder.

This is a feature to create card like images which can be shared on social media for creating awareness.

Additional to the search functionality, a user can get an info graphic (like the one shown below) when he or she search for a specific word. It can be downloaded as an image (worst case, take a screenshot) and can be shared further.

If anyone has any idea how this can be done and wants to take it up please let me know here. So that we can discuss further.

make a google/zoho or any other form(s) which can be used to collect -

• missing words in database
• (not related to this database but) any other incorrect usage of hindi/english words/grammar in marathi.

• and to add a link to these forms from the home page.

Program to take a database (csv format currently) as input, keep only the necessary data (as per a filter criteria which is another input), and output this data in the same format as the input database.

We are looking into the possibility of adding a dedicated database instead of just a csv file. For this it will be handy to have a script that converts csv into a more program friendly format like .sql file or a .json file.

this can be used to link to the "read more" section on the website

in below priority -

1. missing marathi words (translations)
2. topics for each word
3. more words (focus on daily use words)

हजारांच्या, लाखांच्या -
सहस्त्रावधी, लक्षावधी, कोट्यावधी, अब्जावधी
असे संख्यावाचक मराठी शब्द आहेत.

वरील शब्द डेटाबेस मध्ये जोडणे.

on running main.py , the intermediate file "filtered.csv" is left as it is after the md files are generated.
This file should be removed OR even better - never be created in the first place and instead the filtered csv "data" should be passed as data structure (like a row list or something).

this is low priority

currently all md files generated using the python scripts has the header row of "en" and "mr" in it.
It (the csv header row) should be removed at the top level function where it is read for the first time.

Currently most of the documentation is in English but I would like to have the Marathi version for each.

So anyone who is good at and interested in creating stuff in Marathi is most welcome.

Marathi will primarily be needed in -

• The repo documentation
• The website pages

• There are some words in the database where English word is present but the Marathi word is absent.
• For such missing Marathi words, we can have a page where all the English words are displayed along with a text box beside it (to enter the missing Marathi word) and press a submit button at the top or bottom of the page to submit the words to the database.
• These user submitted words will be collected in another database and will enter the main database upon review.

csv contains "experience"

• searching "Experience" doesn't give any results.
• searching "experience" gives expected results.

आपले instagram पेज आहे का? नसले, तर बनवूया. Reddit पेक्षा instagram वर जास्त लोक असतात.

• the db file needs to be sorted using the english word column
• sort order will be A at top and Z at bottom

• All textual content on the home page needs to be translated to Marathi.
• Both Marathi and English text to be kept on the website

The topics list file is currently manually created.

It should be created using a script which uses the topicsParser script.

Please use the develop branch for development.

On the search site, currently the user needs to enter the word in the textbox and then click on "search" button to actually begin the search function.

This can be improved slightly by running the search function directly when users presses the "Enter" key on his keyboard.

• make sure that all rows have the correct number of elements separated by commas
• use the top row as a reference to decide the number of elements

A web app so that layman users can search for words that they need, or adding new words could be simplified.

Possible Problem:
Everytime a new english word has to be added we have to make changes to csv file and open a PR. Even if in future a github.io page is added, it will still be diffiicult to work with a single csv file.

Proposed Solution:
Web app with its database which can be used to fetch words through a Github page (frontend). And this app can also provide a simple UI to add new English words and their alternative Marathi words.

• all pages to have same styling
• for now, use the home page style as a reference. (No explicit styling has been applied there. It's the default style without any theme being applied.)
• static pages must be able to be viewed without any issues even if the user turns on "reader mode" on their browser.

• Users should be able to submit new words
• Submission should happen on a dedicated page/form
• The submitted words should get added to a database of "user suggested words" and not the main database. From there they will be reviewed by the maintainer and added to the main database.
• the main database is currently in a simple csv file, but any other suitable type of database (of user suggested words) for this feature is ok.

on the search page, entered "experience" in the text box.

output was -

Results:
experience : अनुभव
English example currently not available for this wordMarathi example currently not available for this word

but the csv contains both the example sentences.

Template file (word-template.MD) is updated in template folder.
@zarbod can you please update the gen block script to output the block in the updated format?

This block basically is updated due to the new columns in the database.