Panic in field array dim calculation

[kossnikita]

EXINT:
  INTEN:
    _array:
      INTEN?,INTEN10:
        description: "Interrupt enable or disable on line %s"

Token boundaries are calculated without comma split.

svdtools/src/patch/register.rs

Line 361 in 27cf2b1

let (li, ri) = spec_ind(fspec);

li = 5, ri = 8
thread 'main' panicked at 'byte index 18446744073709551614 is out of bounds of `INTEN0`', src\patch\register.rs:383:30
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

svdtools/src/patch/register.rs

Line 383 in 27cf2b1

.map(|f| f.name[li..f.name.len() - ri].to_string())

[burrbull]

I know about this issue, but I don't know good way to fix it.

[burrbull]

#162 should help for your specific situation, but this is just a patch.

[burrbull]

Although this can break other use cases.

[kossnikita]

Yes, works in this case

EXINT:
  INTEN:
    _array:
      INTEN?,INTEN1[0-7]:
        description: "Interrupt enable or disable on line %s"

<fields>
  <field>
    <dim>18</dim>
    <dimIncrement>0x1</dimIncrement>
    <dimIndex>0-17</dimIndex>
    <name>INTEN%s</name>
    <description>Interrupt enable or disable on line %s</description>

But don't work there (I don't know why someone would write such a weird patch)

EXINT:
  INTEN:
    _array:
      INTEN1[0-7],INTEN?:
        description: "Interrupt enable or disable on line %s"

<fields>
  <field>
    <dim>18</dim>
    <dimIncrement>0x1</dimIncrement>
    <dimIndex>,,,,,,,,,,0,1,2,3,4,5,6,7</dimIndex>
    <name>INTEN1%s</name>
    <description>Interrupt enable or disable on line %s</description>

[burrbull]

I don't have better ideas.

P.S. We need someone to improve YAML Format as patch format was much extended since last docs edition. Do you have a wish to help? I'm bad in doc writing. (#5)

[kossnikita]

As far as I know, the functionality of rust and python is now slightly different. Should this be reflected in the documentation or should the new syntax be introduced into Python first?

[burrbull]

Python version is deprecated, but we try to support it if it is not hard. For example #162 is easy, but I can't say the same for #159 (the magic that it needed is in svd-rs). For now differences are not big.