cdsoft / panda Goto Github PK

I'm getting an error whenever I have a boolean value in my metadata, for example link-citations: true

I believe the error is coming from the read_vars_in_meta function, when it tries to unpack the table.

This works:

::: {.if myTest=test}
Using the syntax from test.md, this should appear because myTest is set to "test".
:::

It renders as:

Using the syntax from test.md, this should appear because myTest is set to “test”.

...but, this does not - it renders the block as plain text:

Here is some text and...
::: {.if myFlag=true}
This should appear because myFlag is set to true.
:::
Oh, and more text.

It renders as:

Here is some text and. . . ::: {.if myFlag=true} This should appear because myFlag is set to true. ::: Oh, and more text.

This greatly diminishes the utility of conditionals.

Thanks for this project! I'm currently evaluating whether I can replace pp with panda or if I still need a preprocessor like upp.

Is there a way to access the name of the current file from panda? For example, with pp I used the macro !exec(date -r !file "+%Y-%m-%d") to set the metadata date to the date of the file's last modification.

For some absurd/historical reason the F# interactive console for scripts requires the name of the script files to end the extension .fsx. That means that something that should work perfectly well like this:

```{.fsx cmd="dotnet fsi" }
open System
Console.WriteLine("Hello from F#!")
```

Simply doesn't work. Instead I get the following error:

"~/Documents/Wiki/src/wasabi/wasabito_compact_filters.md" 50L, 729B written
^@^@error FS0226: The file extension of '/tmp/panda_script-57a468d967123abc/script.dotnet' is not r
ecognized. Source files must have extension .fs, .fsi, .fsx, .fsscript, .ml or .mli.^@^@Error runni
ng filter /home/lontivero/.local/bin/panda.lua:^@PandocLuaError "/home/lontivero/.local/bin/panda.l
ua:434: script error"^@stack traceback:^@^I/home/lontivero/.local/bin/panda.lua:422: in upvalue 'ru
n_script'^@^I/home/lontivero/.local/bin/panda.lua:446: in function </home/lontivero/.local/bin/pand
a.lua:440>^@
Press ENTER or type command to continue

I don't see any obvious way to make panda to save the script with a given extension.

Just to prove the problem is the script file extension, if I hardcode th extension .fsx in the run_script function as follow:

It works:

When included file is a table, the caption must be a part of the included file or else the caption is not recognized. For example if table.md contains:

Col A|ColB
-----|-----
1|2

...and the following is specified in the main file:

:::{include table.md}
:::

Table: This is my Table

The caption will not render as a table caption. For that to work, table.md must contain the following:

Col A|ColB
-----|-----
1|2

Table: This is my Table

With the main file specifying only:

:::{include table.md}
:::

The problem with this is that if one is attempting to convert a pipe-delimited text file from Excel or some other input source,
neither the header separator (i.e. "-----|-----") nor the Table caption will be applied. So it would be extremely useful if one could
use a table.md file which looked like this:

Col A|ColB
1|2

...and state in the main file:

:::{include table.md caption="This is my Table" header=true}
:::

So that the result would render as:

Table: This is my Table

e.g. "Echo Hello World{cmd=cmd.exe}" or "Echo Hello World{cmd=cmd.exe /c}"

Variable expansion does not work if there are punctuation characters immediately to either side of the {{variable}} reference. This makes it impossible to use {{variables}} at the end of sentence, for example. See attachment below.

Also, conditional syntax as described in the rendered documentation does not match actual conditional syntax appearing in test.md in the repo.

bugs.md

i.e., the following:

@startuml
Alice -> Bob: hello
@enduml

...should result in this:

![Alice & Bob](.build/cache/0ec62f1568ac33e20ec8d430ae77a9cbe6c9cd46.svg "Alice & Bob")

...rather than this:

![](.build/cache/0ec62f1568ac33e20ec8d430ae77a9cbe6c9cd46.svg "Alice & Bob")

While graphviz diagrams work well, gnuplot doesn't. I've tried with all possible ideas but nothing seems to work. This example is taken from the doc (with img and out modified). I have gnuplot installed

```{render="gnuplot -e 'set terminal svg' -e 'set output "%o.svg"' -c %i"
     img="../../src/images/panda_gnuplot_example"
     out="/home/lontivero/Documents/Wiki/src/images" }
set xrange [-pi:pi]
set yrange [-1.5:1.5]
plot sin(x) lw 4, cos(x) lw 4
```

This is the result (pandoc rendered html output):

I don't know how to debug nor how to get more info to help understand what can be going wrong. Any idea?

It is sometimes useful to highlight a line or section of code inside a code block in order to bring attention to it. Is this something that panda would consider implementing?

For example

```{highlight=2-3}
line1
line2
line3
```

might result in lines 2 and 3 being hilighted

I'm using panda as part of my wiki site generator and i need the paths to be relative to the wiki root directory. I wanted to solve it by doing:

```meta
wroot = "/home/user/Documents/Wiki/src/wasabi"
```

:::{include={{wroot}}/radical_ideas.md}}
whatever
:::

But it doesn't include the file. Instead it results in:

However, if I hardcode the file path it works like charm.