make the syntax compatible with pandas

[arita37]

Thansk for good work !

in order to increase user based
and facilitate user migration,
please make the syntax one to one match with pandas:

Pd.read_csv(…)

check koalas library

[hkpeaks]

you can see this example

To support Python:

import peaks as hk

Table = hk.ReadFile("10MillionRows.csv")
Table2 = hk.Distinct("Table | Ledger, Account, PartNo,Project,Contact,Unit Code, D/C,Currency")
FilePath = hk.WriteFile("Table2 | * ~ Peaks-Distinct10.csv")
FilePath2 = hk.WriteFile("Table | Ledger, Account, PartNo,Project,Contact ~ Peaks-Transaction.csv")

where Table is variable which contain a in-memory table, FilePath is a string variable
File type is auto detected by .ext and its bytearray, so no need _csv

To support Web request and Windows/Linux command line (can be ran by script and command line):

ReadFile{10MillionRows.csv ~ Table}
Distinct{Ledger, Account, PartNo,Project,Contact,Unit Code, D/C,Currency ~ Table2}
WriteFile{Table2 | * ~ Peaks-Distinct10.csv}
WriteFile{Table | Ledger, Account, PartNo,Project,Contact ~ Peaks-Transaction.csv}

where the Python Code (" ") is equvalent to the original syntax {}.
And "variable =" is equvalent to the original syntax "~ TableName"
When output table is a file name instead of in-memory table, the variable
will be a string which contain a full file path of the output file.

[arita37]

Ok, method names are not matching pandas…. It’s a pain for people to re-learn method names. What polars, koalas are sucessful because they match pandas naming Before its too late, advise to do it…. But its your library and you decide !!! Anyway, keep the good work !

…

On Apr 22, 2023, at 11:03, Max Yu ***@***.***> wrote:  you can see this example To support Python: import peaks as hk Table = hk.ReadFile("10MillionRows.csv") Table2 = hk.Distinct("Table | Ledger, Account, PartNo,Project,Contact,Unit Code, D/C,Currency") FilePath = hk.WriteFile("Table2 | * ~ Peaks-Distinct10.csv") FilePath2 = hk.WriteFile("Table | Ledger, Account, PartNo,Project,Contact ~ Peaks-Transaction.csv") where Table is variable which contain a in-memory table, FilePath is a string variable To support Web request and Windows/Linux command line (can be ran by script and command line): ReadFile{10MillionRows.csv ~ Table} Distinct{Ledger, Account, PartNo,Project,Contact,Unit Code, D/C,Currency ~ Table2} WriteFile{Table2 | * ~ Peaks-Distinct10.csv} WriteFile{Table | Ledger, Account, PartNo,Project,Contact ~ Peaks-Transaction.csv} where the Python Code (" ") is equvalent to the original syntax {}. And "variable =" is equvalent to the original syntax "~ TableName" When output table is a file name instead of in-memory table, the variable will be a string which contain a full file path of the output file. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.

[hkpeaks]

Need to provide a clear naming, and I will request Pandas and Polars to follow the spec of Peaks Framework, to standarsize ETL expression similar to HTML5. And there are room different software developers joining together to create a standardize ETL expression. If only command ReadFile -> Read obviously not an issue, but there are a lot of other ETL expressions. You can have a look on this doc "WebNameSQL.pdf".

[arita37]

Ok, good luck. You lost me as a user and probably million of others… 1) nobody use csv to store data your benchmark makes no sense 2) Prefer using Polars, makes no sense to use another crappy library who dont care about users

…

On Apr 22, 2023, at 13:58, Max Yu ***@***.***> wrote:  Need to provide a clear naming, and I will request Pandas and Polars to follow the spec of Peaks Framework, to standarsize ETL expression. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.

[hkpeaks]

CSV is one of the formats that any dataframe software supports. Other formats include Parquet, Excel, JSON, HTML and more. However, this software is not intended to be another Polars or Pandas. The benchmark test will mainly focus on some of the most common ETL functions such as ReadFile, Select, Distinct, GroupBy, JoinTable, OrderBy, AddColumn, Crosstab and WriteFile. Other functions will be significantly different from Polars or Pandas. Peaks is derived from my previous experience of accounting software and will be extended to support bioinformatics.

You should be an excellent user. Polars is one of my best recommended dataframe software.

[arita37]

No interest in your software… And not anymore to promote Your lost people Good luck

…

On Apr 22, 2023, at 18:07, Max Yu ***@***.***> wrote:  CSV is one of the formats that any dataframe software supports. Other formats include Parquet, Excel, JSON, HTML and more. However, this software is not intended to be another Polars or Pandas. The benchmark test will mainly focus on some of the most common ETL functions such as ReadFile, Select, Distinct, GroupBy, JoinTable, OrderBy, AddColumn, Crosstab and WriteFile. Other functions will be significantly different from Polars or Pandas. Peaks is derived from my previous experience of accounting software and will be extended to support bioinformatics. You should be an excellent user. Polars is one of my best recommended dataframe software. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.

[hkpeaks]

Polars is one of excellent software for you. I shall introduce it for more friends and companies. In fact I have learnt a lot from Polars. I enjoy using it for testing. Previously I reported issues for billion rows testing, subquently the author fix it.