big update for doshshell.

hello everyone, today I introduce to you a shell I have been developing for a while now and have just come back

and added features to it.

I have to share a few reasons for building it.

I primarily use Windows as my dev environment and most of the time I work with PowerShell. PowerShell is powerful (it’s Microsoft after all) but

its syntax is too complicated and hard to remember. Also, PowerShell focuses on performance so modern shell features like fishshell

are not available. I once wished Windows could run bash because working with PowerShell was too boring. I also tried to find ways to install bash on

Windows, but there were only instructions to install it on WSL or Git Bash. Feeling frustrated, I decided to build my own shell for use. Since I went through the trouble of building it

I made it cross-platform

I think that on all platforms, running scripts or commands should just be one way instead of having each OS run differently like now

therefore doshshell was launched to do just that. No matter what OS you are on, it runs in the same way, helping everyone work in sync with the shell on all platforms

the second goal of doshshell is to replace traditional shells that are sluggish, slow, feature-poor, and hard to configure like bash and zsh

with a shell that has simple syntax, flexibility, speed, and ease of configuration

and the most important difference is that doshshell differs from traditional shells in that, I learned a very nice design philosophy from nu.

For bash/zsh, "Everything is text" means all pipelines are text so they easily break when the output format changes.

Whereas dosh, by design, follows "Everything is structured data". Each pipeline fully supports data types

String

Number

Boolean

Record

List

Table

DateTime

Duration

Filesize

unlike just text, text, and text of bash/zsh

- so what are its advantages??

- if the pipeline is text, you have to manually parse data, but if each pipeline represents clear data types

then the subsequent pipeline understands what the previous one is, what it returns. Thus, operations like searching or querying data are very

quick and convenient.

for example, getting a list of files you need looks like this in bash/zsh:

```

find . -maxdepth 1 -type f -size +1k \

| xargs stat --format="%s %n" \

| sort -n

```

It seems concise, but the syntax is quite confusing, making it hard to understand

with dosh, it’s easier and clearer

```

ls | where size > 1kb | where type == file | reverse | sort-by size

```

at a glance, you can understand the function of each pipeline.

or another example. For the process: find all .xlsx files, open the Users sheet, find the cell containing the text DoShell, change it to Dosh, and save to a new file in the out/ directory.

in dosh:

```

glob "**/*.xlsx"

| open-each

| sheet "Users"

| replace "DoShell" "Dosh" --recursive

| save --to-dir out --suffix ".updated"

```

Or more clearly:

```

glob "**/*.xlsx"

| open-each

| sheet "Users"

| where contains "DoShell"

| replace "DoShell" "Dosh" --recursive

| save --to-dir out --suffix ".updated"

```

Desired result:

from: reports/users.xlsx

to: out/users.updated.xlsx

with bash, it cannot process .xlsx, so you have to