1492678 - XDR-encode/decode the source text of scripts using codeChars for better typing and to address endianness considerations

Status: RESOLVED FIXED

(Core :: JavaScript Engine, enhancement)

Comment 1

[Jeff Walden [:Waldo]]

Attachment: Patch

Comment 2

[Ted Campbell [:tcampbell]]

Comment on attachment 9010503 [details] [diff] [review]
Patch

Review of attachment 9010503 [details] [diff] [review]:
-----------------------------------------------------------------

Much improved, thanks.

::: js/src/vm/JSScript.cpp
@@ +2230,5 @@
>      {
>          size_t match(Uncompressed&) {
> +            // Return 0 so the |if (compressedLength)| tests below correctly
> +            // (and differently) XDR scripts with compressed source and scripts
> +            // with uncompressed source.

// Return 0 for uncompressed source that |if (compressedLength)| tests below can differentiate between compressed and uncompressed sources.

@@ +2277,2 @@
>  
>          uint32_t compressedLength;

// A compressed length of 0 indicates source is uncompressed.

Comment 3

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/cb4aa4bcb7fa
XDR-encode/decode the source text of scripts using codeChars, not codeBytes that can scramble endianness of char16_t source text if the XDR data is read back on a machine of different endianness.  r=tcampbell

Comment 4

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/cb4aa4bcb7fa