Re: [GT] Input validation in Tools.pm

["Robert A. Schmied" <ras AT acm.org>]

sam wrote:
> Thx for answer Robert.
> My issue with I:STO was not warnings
> generated by the --full, I was getting absolutely not output at, and
> same thing with display_indicator.

yea! no output is another side effect of the trunk version of display_indicator.
it really could use some rework, but for indicator development + test + evaluation
it's not too bad. i never use --full due to the (potential) lack of initial
dependencies problem, besides from my experience market conditions more than
a few years back don't affect the current market much. your experience may differ.

do you use it (display_indicator/display_signal) to generate raw data processed by other programs?

> Anyway, after an rm/cvs update, it's fine, with the trunk version.
> I'm a bit lost, what is the difference between the "exp" branch and "experimental-modules" branch ?

thomas weigert has been making many changes and is doing so in the "branches/exp" branch,
this to avoid killing the working "trunk" branch.

the "branches/experimental-modules" branch was used a long time back to migrate to svn
or maybe rearrange the code tree. the "experimental-modules" branch is strictly
don't use for anything, ever.

> And I suppose your example, with --ti and --sta is your private hacked version, in your personnal repo ?

humm -- i believe that Getopt::Long (at least from v 2.33 onward) will accept option flag
abbreviations that are unique in the set of option flag names. but there might be a config
setting for Getopt::Long that i turned on, but now fail to remember or find any mention of.

--ti is unique for --timeframe and --sta unique for --start as least with display_indicator flag names.

you should be able to easily validate this by trying it. let me know (via list is fine)
if your trunk install fails to accept these abbreviations and i'll look harder for it.


> I've not found any STO2, just found STOkw, which seems to be now the "official" STO indicator.

looks to me like the current trunk STO has all the improvements contributed by
karsten and thomas. it looks like what i called STO2 locally during t+e.

i think STOkw has been depreciated in favor of the latest STO.

> For
> the LinearRegression, there is some information available in wikipedia
> and investopia.com, for example.  But didn't find a link which explains
> clearly why it's important in TA.

yep -- that's about where i am -- not being able to figure out what to do with it.

> Sam.

<snip>