0.5.0

breadroll is now in public beta. We are super excited to see breadroll in the wild and even more excited to see contribution from the future breadroll community. So many thing are packed into the new breadroll. We have leveraged TypeScript's static typing to prevent data-related runtime errors, improved developer experience, improved API design, and added a few more powerful methods.

• A brand new parser experience utilizing mafintosh/csv-parser (opens in a new tab) via sindresorhus/neat-csv (opens in a new tab) see the former for the default behavior of the parser.

• Now breadroll Dataframes have types; assign types by using the .open methods like so; Breadroll.open.local<T>() or directly like so; new Dataframe<T>()

• We added more methods to the Dataframe class

  • .concat
  • .merge
  • .shape
  • .tail
  • .toNumber

• We modified the Breadroll.open methods to have a dedicated seperator or delimiter function parameter, this way we reduce the number of instances need to open multiple file with different formats; eg. Breadroll.open.local<T>("./path/to/file.csv", ",").

  This simply means we've moved the delimiter from Breadroll.DataframeReadOptions to each individual method.

• breadroll being in public beta means you get to build the future of breadroll, want to fix a bug, request a feature, improve testing; now is the time - take a quick glance on how to get started with contributing to breadroll

• 67 🌠 stars on GitHub, thank you so much for the support - help breadroll rise even higher, drop a star.

0.4.0

In breadroll's new version 0.4.0 a lot of new shiny features are introduced, Here are the cool new features

• Improvements to the Breadroll and the Dataframe API
• Introduction of a new remote data source
• Changed filter keywords in favor of symbols over length words
• A snazzy new way to perform data transformation
• Introuducing NumericConstants - a collection of mathematical and physical constants
• breadroll hit 35 🌠 stars on GitHub - help breadroll rise higher, drop a 🌠 star.

API Improvements

The Breadroll & Dataframe APIs got a little face lift, with two major addition to the Breadroll API and a few more others to the Dataframe API;

• DataframeReadOptions?.parseNumber: boolean - This is an optional property that allows you to opt out of parsing numbers defaults to true, when set to false all values are returned as strings
• DataframeReadOptions?.supabase: SupabaseClient - This is an optional property takes in the Supabase client created via createClient(..., ...), this property enables Breadroll.open.supabaseStorage method.
• By default headers (labels) were being converted to lower-case, now headers (labels) maintain case-sensitivity
• Dataframe.apply({...}: Apply) - This function allows you to perform data transformation on a specified column of the dataframe

New Remote Data Source 🚀 - Supabase

• Breadroll.open.supabaseStorage(bucketName: string, filepath: string) - This function takes in the bucket name and filepath of the file, downloads the file, parses the files and retuns a Dataframe

Symbol-based Filter Keywords

The use of symbols as keyword ensurces less verbosity; most of the filter keywords have been replaced with symbols; "eqauls" is now "==" see the full changes here

Numeric Constants

In addition to having the ability to perform data transformation using Dataframe.apply({...}), the numeric constant object provides mathematical (math) and physical (physical) constants. These two can be nicely paired to make cool data transformation, most especially with scientific datasets.

These values are double-precision 64-bit binary format IEEE 754 value, which have ±15 decimal digits of precision. The physical constant are provided in their standard SI units

0.3.7-alpha

In breadroll's new version 0.3.7-alpha two new features are introduced, Here are the cool new features

• Regex filter using "matches" keyword
• Integer-based indexing / selection

Regex filter

Introducing Regex filter with the keyword "matches" for performing complex queries like matching certain patterns in a String, this uses the matches keyword and takes in a RegExp (opens in a new tab). We recommend using this filter when a trade off on time is acceptable, see here for a more detailed explanation.

const filtered = selected_cols.filter("job_title", "matches", /engineer/i);

Integer-based indexing / selection

Dataframe.rows({start?: number, end?: number}: Indexer): Dataframe Returns a specific number of rows of the dataframe, this function takes an object with either the start or the end index and return the dataframe with the specified number of rows - check it out here

const count: Dataframe = df.rows({ start: 0, end: 4 });

Dataframe.cols({start?: number, end?: number}: Indexer): Dataframe Returns specific columns of the dataframe strictly using interger based indexing similar to panda's iloc - check it out here

const count: Dataframe = df.cols({ start: 0, end: 4 });

0.3.6-alpha

Breadroll.open.https(url: string, headers?: Headers): Promise<Dataframe> This function fetches and return a file via a URL over https, with a default GET method, with optional provision for custom headers

0.3.1

Dataframe.select(keys: Array<string>): Dataframe This function return a new Dataframe with only the desired rows, ie. rows with the specified labels

0.3.0

Dataframe.to_blob(filetype: "csv" | "tsv"): Blob This function converts the current dataframe into a Blob of MIME filetype "csv" | "tsv"

Dataframe.save

• .json(filepath: string): This function saves Dataframe as a JSON files ie. for example Dataframe.save.json(filepath: string).
• .csv(filepath: string): This function saves Dataframe as a CSV files ie. for example Dataframe.save.csv(filepath: string).
• .tsv(filepath: string): This function saves Dataframe as a TSV files ie. for example Dataframe.save.tsv(filepath: string).

0.2.0

Dataframe.min(key: string): number This function returns the minimum value of all the values of the specified column ie. key. Note, the values are coerse into a number type

Dataframe.max(key: string): number This function returns the maximum value of all the values of the specified column ie. key. Note, the values are coerse into a number type

Dataframe.sum(key: string): number This function returns the sum of all the values of the specified column ie. key. Note, the values are coerse into a number type

Dataframe.average(key: string): number This function returns the average of all the values of the specified column ie. key. Note, the values are coerse into a number type

0.1.3

Dataframe.use(callback: (dataframe: Array<Record<string, unknown>>) => Dataframe): Dataframe it provides access to the object ie. it kinda ejects from the base class allowing user to perform their own custom operation on a the current dataframe version, eg. after running Dataframe.filter

Added more filters, greater than or equal to, less than or equal to, is between

Tweaked Dataframe.filter to take a fourth and optional argument limit for range filters like is between

0.1.0

• Dataframe.head: Dataframe returns the first five rows of the dataframe
• Dataframe.labels: Array<string> returns the labels of the dataframe
• Dataframe.filter(key: string, filter: Condition, value: unknown) - filters the rows in Dataframe and returns Dataframe(dataframe)
• Dataframe.value - exposes the Dataframe's dataframe as Array<Record<string, unknown>>
• Dataframe.count - returns the count of rows in the dataframe or dataframe object
• Dataframe.isNull - returns all the rows that have some properties equal to null
• Dataframe.notNull - returns all the rows that have every property equal to !null
• Dataframe.dtypes - returns all the data types of all the columns in the dataframe