Can MAX_REFLOW_DEPTH be increased?
In the days of Windows 95, recursive algorithms in layout used to overrun the call stack on Windows unless the depth of the DOM tree was limited. The HTML parser still enforces the limit, and people complain about it from time to time. (Most of the time, the Web pages that hit the limit are doing something wrong, but that's normal.) Does layout still use recursive algorithms that require a depth limit? If the depth limit is still needed, can now be increased? (The limit is defined at https://mxr.mozilla.org/mozilla-central/source/layout/generic/nsIFrame.h#16 ) -- Henri Sivonen hsivo...@hsivonen.fi https://hsivonen.fi/ ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: ICU proposing to drop support for WinXP (and OS X 10.6)
On Mon, May 2, 2016 at 2:37 AM, Jim Blandy wrote: > What are the distributions of memory and flash sizes for the devices people > currently run Fennec on? It'll be almost impossible to have a good > discussion about Fennec size without those numbers. I seem to remember that > is data we felt was okay to collect. We should also be data-driven about understanding where the bytes go. In particular, I think functionality-neutral size reductions should be done before blocking new functionality from landing. In addition to unoptimally compressed PNGs, there's unminified JS in Fennec (i.e. the JS comments are shipped). -- Henri Sivonen hsivo...@hsivonen.fi https://hsivonen.fi/ ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Can we remove nsIEntityConverter?
On Mon, May 2, 2016 at 9:45 AM, Frédéric Wang wrote: > Le 01/05/2016 02:16, smaug a écrit : >> What would source view for mathml look like if we removed >> nsIEntityConverter? > > AFAIK, the only point is to replace things like "∑" with "∑" > in order to make it more readable. However, with appropriate fonts > installed I think reading "∑" is also fine. Yes, and if someone wants to special-case and , special-casing two characters takes less space than the entire entity tables. As for the size question, the data tables take 2336 bytes in omni.ja plus the zip directory overhead plus the C++ code that enables the use of these tables. > Le 30/04/2016 12:25, Henri Sivonen a écrit : >> In desktop Firefox, these data tables are used only for the >> MathML View Source feature. > > Personally, I don't really use this feature, as I find the DOM inspector > or the "MathML Copy" add-on (*) more convenient to check or copy a > MathML formula. > > I guess we can move this feature from the Desktop front-end to a > separate Add-on (that could potentially work on mobile too in the > future). However, I can't speak for the users. Maybe we should write to > the Math WG mailing list in order to get more feedback. > > (*) https://addons.mozilla.org/en-US/firefox/addon/mathml-copy/ Even if someone uses the current feature, special handling for examining MathML seems like something that should be in an extension. -- Henri Sivonen hsivo...@hsivonen.fi https://hsivonen.fi/ ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Can MAX_REFLOW_DEPTH be increased?
On Mon, May 2, 2016 at 10:22 AM, Henri Sivonen wrote: > If the depth limit is still needed, can now be increased? What do other UAs do? This seems like it may as well be standardized, just so the odd corner-case page breaks the same in all UAs. ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Can MAX_REFLOW_DEPTH be increased?
On Mon, May 2, 2016 at 2:01 PM, Aryeh Gregor wrote: > On Mon, May 2, 2016 at 10:22 AM, Henri Sivonen wrote: >> If the depth limit is still needed, can now be increased? > > What do other UAs do? This seems like it may as well be standardized, > just so the odd corner-case page breaks the same in all UAs. Whenever people complain about this, the pages render "correctly" in at least Blink and WebKit. I don't recall checking Edge. -- Henri Sivonen hsivo...@hsivonen.fi https://hsivonen.fi/ ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Can MAX_REFLOW_DEPTH be increased?
Test-case:
var cur = document.querySelector("script");
var depth = -1;
while (cur) {
cur = cur.parentNode;
depth++;
}
document.body.textContent = depth;
Outputs 513 in Chrome. It looks like any further s are inserted
as siblings instead of children. The is processed correctly.
IE11 outputs "1003" but then the tab seems to crash.
We don't execute the script at all. This doesn't seem desirable even
if we do limit the depth. A modified test-case for us: