Astro-App-Satpass2
view release on metacpan or search on metacpan
unable to access that from a macOS launchd job, even running as the
owner of the direcory. At the moment I am also looking in
File::HomeDir->my_dist_data(), which experimentation seems to show
is accessible to a launchd job. But that may not be its final
resting place. The my_dist_config() location will still be found --
this is just a band-aid for those who use this code in scheduled
jobs.
0.043 2020-01-17 T. R. Wyant
Fix y2020 problem when using Date::Manip for input. The ISO8601
parser appears to be unaffected.
0.042 2019-11-03 T. R. Wyant
Fix redirects in POD URL links.
Implement \U and friends inside double quotes, except for \Q. If run
under Perl before version 5.15.8, \F is equivalent to \L.
Add interactive method error(). This aborts the macro or include
file in which it appears, displaying an error message.
Add if() operators -z and -n, which are analogous to the test (1)
operators of the same name.
Macro definitions are no longer unescaped. Unescaping was preventing
the \U ... functionality from working in a macro, and had no known
benefits. I must have put it in there for some reason, but I no
longer know why.
Correct eg/pass_ics to alert before and at rise.
0.041 2019-10-04 T. R. Wyant
Fix validation error caught by Astro::Coord::ECI 0.109_01.
Validate horizon, latitude, and longitude.
Add -verbose to macro( load => ... )
0.040 2019-07-20 T. R. Wyant
Fix broken POD links and add test to ensure they stay fixed.
Remove Astro::App::Satpass2::Format::Classic refs. This module was
retired quite a while back.
0.039 2019-03-29 T. R. Wyant
Fix test failures when local zone not determined. This was a single
test in t/format_template.t that attempted to use
DateTime::Calendar::Christian to format March 15, 44 BC. The problem
is referred from DateTime::TimeZone::Local, which can not determine
the local zone under Haiku, and dies. Thanks to Serguei Trouchelle
for the tests that exposed this.
0.038 2019-03-13 T. R. Wyant
The 'model' setting is no longer ignored. It is now actually applied
to the TLE objects when loading them, and a change in value causes
all loaded TLEs to be modified as well.
Make ::TLE::Iridium optional.
0.037 2018-10-28 T. R. Wyant
Fix test failure in t/locale.t. I hope. Some of the CPAN testers
get errors like
do "t/locale/fu_BAR.pm" failed, '.' is no longer in @INC; did you
mean do "./t/locale/fu_BAR.pm"? at ...
gglib/Astro/App/Satpass2/Locale.pm line 210.
but I can't duplicate this, even under Perls that manifestly do not
have '.' or any facimile thereof in @INC.
0.036 2018-10-17 T. R. Wyant
Support alternate Sun and Moon classes, and other bodies that
contain built-in models (e.g. Astro::Coord::ECI::VSOP87D::*).
Add sky() settings to the output of save().
Proper display of formatter template names.
0.035 2018-01-26 T. R. Wyant
TODO test because of changed equinox algorithm.
Remove mailto: bug reporting metadata. Not that I refuse these, it
is just that it appears to cause RT to think I will not accept bug
reports from them.
Support for decayed Iridium satellites.
0.034 2017-04-01 T. R. Wyant
Update FormatTime/DateTime.pm calendar_name code to take advantage of
the calendar_name() method if present.
Add 'provides' data to ExtUtils::MakeMaker output
0.033 2017-01-12 T. R. Wyant
Protect tests from local zone problems
Use 'use parent' instead of 'use base'.
0.032 2016-12-11 T. R. Wyant
Use time_local() and time_gm() from Astro::Coord::ECI::Utils. This
requires Astro::Coord::ECI::Utils 0.077.
The problem here is that earlier in the year I found that Time::Local
worked better for dates well away from the current year if I passed it
the Gregorian year rather than computing the Perl year. But I support
Time::y2038 for the benefit of older Perls, and it expects a Perl year.
The fix was to wrap the Time::y2038 calls in code that duplicates
Time::Local's year handling.
Thanks to Slaven Rezic for finding this, and for having smokers
configured with modules like Time::y2038 (which is optional for this
package) so that he _could_ find it. Kudos!
Better diagnostics for Date::Manip::DM5 test fails
Fix deep recursion in ::Macro->wail() & friends
Support formatting times with any class conforming to the DateTime
interface, via formatter attribute 'back_end'. If
DateTime::Calendar::Christian is used her, require version 0.06.
Support using DateTime (if available) to convert a parsed ISO-8601
time to epoch. An arbitrary class confirming to the DateTime
interface can be used instead, if specified.
and document its arguments.
The formatter object now displays its time_formatter attribute as a
short name where possible.
Add time_formatter value to 'show' output.
Allow specifying a time as a scalar reference, The dereferenced value
is interpreted as epoch.
Special-case DateTime strftime() format %{calendar_name} to display
the name of the calendar in use.
Heuristics to make ODF geocoding more useful. I hope.
Retract support for Geo::Coder::Geocoder::US. The underlying web site
has been AWOL since late 2015.
0.031 2016-02-24 T. R. Wyant
Consolidate development changes into next production release.
Correct and clean up recommended-module code.
Document anticipated retraction of Geo::Coder::Geocoder::US geocoding
support. Skip t/geocode_geocoder_us.t unless author testing
Add Makefile targets authortest, testcover
0.030 2016-01-05 T. R. Wyant
Add GitHub repository to metadata.
0.029 2015-12-23 T. R. Wyant
Make OSM the default geocoder.
The problem is that geocoder.us (the originial geocoder) has been
offline for a couple weeks, and it makes no sense to have a broken
default.
0.028 2015-08-29 T. R. Wyant
Remove notices saying this is alpha code.
Try to quash errors under Cygwin using DateTime. The problem is that
we think we can use it, but we get failures when initializing
because DateTime::TimeZone can not determine the local zone. We
handle this by checking for it and falling back to POSIX formatting
if needed.
Sanitize TZ before using it in tests.
Eliminate use of deprecated Geo::WebService::Elevation::USGS 'source'
attribute.
0.027 2015-07-11 T. R. Wyant
This release should be functionally identical to 0.026, but was made
because I wanted a production release that superseded 0.026_01,
which was made due to a bad patch I developed for Date::Manip.
0.026 2015-06-26 T. R. Wyant
Change localization of equinoxes and solstices to take advantage of
the Astro::Coord::ECI southern hemisphere functionality. This
includes not only the default C locale but eg/es.pm.
Specify encoding UTF-8 for source files. This was always intended, but
never actually done.
Skip t/parse_time_date_manip_6.t if Date::Manip and DateTimeTimeZone
come up with different zones. This is because Date::Manip may not
default to the right zone under at least some BSD variants,
resulting in failing tests.
0.025 2015-03-29 T. R. Wyant
Track interface change to Date::Manip >= 6.49. This involves using the
SetDate configuration variable instead of TZ, and silently making
'gmt' (any case) into 'UT'.
0.024 2015-01-10 T. R. Wyant
Remove TomTom support.
As of January 5 2015, Geo::Coder::TomTom requires an API key. Since
I have not been supporting geocoders that require registration,
TomTom has been dropped. It has not worked since November 2014
anyway.
0.023 2014-11-17 T. R. Wyant
Address CPAN Testers failures in t/locale.t under the combination of
Perl 5.20.1 and MSWin32.
0.022 2014-11-08 T. R. Wyant
Remove all reference to the 'lit' attribute. This has been removed in
favor of 'edge_of_earths_shadow'.
Suspend support for geocoding via TomTom. It appears that TomTom has
retracted the undelying web service. Support in this package will be
retracted May 1 2015 unless the situation is resolved favorably
before then. If the situation is resolved unfavorably before then,
support will be retracted when I become aware of the fact.
0.021 2014-09-17 T. R. Wyant
Correctly handle tz attribute when the DateTime date formatter is in
use.
Fix bug in spacetrack -verbose search_oid and friends.
Add -setup option to perl() method. This causes the arguments to be
recorded and emitted by the save() method.
More informative DateTime time zone error message.
Rewrite object-specific formatter code to place dispatchers in the
Astro::App::Satpass2::FormatValue name space rather than use
AUTOLOAD(). The AUTOLOAD() implementation caused all sorts of
smoker errors that I was unable to reproduce, and I was averse to
blindly writing stubs for Perl-defined subroutines.
Remove subroutine Astro::App::Satpass2::Utils::fold_case(). This is
documented as private, but since is _was_ documented I decided to
put it through an abbreviated deprecation cycle.
Add add() method to Astro::App::Satpass2. This validates and adds TLE
objects.
Remove ::Format::Template::Provider from t/basic.t
Fix test failures in t/locale.t
Force stringification of Template::Exception.
Have the tle_verbose template ignore non-TLE data.
Fix bug in -choose foo,bar which caused only 'foo' to be chosen.
Add real localization support. This addresses almost all text seen by
the user, though I have not done a lot on error messages other than
put in the calls. A very rough Spanish localization is provided in
eg/.
Generate format selection options for tle() based on available
templates. Basically, if you load template eg/tle_json.tt as
'tle_json', the -json option will cause that template to be used.
Drop Astro::App::Satpass2::Format::Template::Provider.
Refactor and rationalize formatter code.
Add Astro::App::Satpass2::FormatValue::Formatter to implement custom
formatters without subclassing. THIS IS EXPERIMENTAL.
Add Satpass2 interactive method perl(). This executes some Perl,
either from a file or via an eval. The invocant comes in in
$ARGV[0].
0.020 2014-06-12 T. R. Wyant
Make use of the TLE illuminated() method To calculate illumination for
the position() method. This requires Astro::Coord::ECI::TLE 0.064.
0.019 2014-05-14 T. R. Wyant
Require Astro::Coord::ECI (etc) 0.063.
Drop default template documentation, which was only intended for
informational purposes. Instead, we document how to extract the
default templates.
Add support for magnitudes. This involves:
* Add a pass_variant attribute so we can ask for the brightest
moment in the pass.
* Add a -brightest option to pass() to override pass_variant, so
that the interface is at least semi-consistent.
* Modify the value formatter's magnitude() formatter so that, if
there is no {magnitude} key in the thing being formatted, but
there is a magnitude() method on the body, that method is called
and the result formatted.
* Modify the template for pass() to check to see whether
PASS_VARIANT_BRIGHTEST is set, and if so add a magnitude column to
the report.
The last point above required a number of supporting changes:
- Add a want_pass_variant() convenience method to return true if the
specific variant is wanted.
( run in 0.532 second using v1.01-cache-2.11-cpan-ceb78f64989 )