lib/sh/util.sh: cd(): support rebasing path vars for the new dir

bugfix: lib/sh/archives.sh: compress(): need to use zip's path syntax to avoid the file in the archive being named "-"

lib/tnrs.py: added option to avoid using TNRS's TSV export feature, which currently returns incorrect selected matches (vegpath.org/issues/943). this has been implemented up through the GWT/JSON decoding.

lib/tnrs.py: added gwt_decode()

lib/strings.py: added unesc_quotes() and helper functions

lib/strings.py: added json_decode()

lib/runscripts/extract.run: export_(): also compress created file

lib/sh/archives.sh: added compress(), expand(), which handle compression of individual files

lib/tnrs.py: documentation about output of the retrieve step: added that this is also unusable because the array does not contain all the columns and contains no column names

fix: lib/tnrs.py: retrieval_request_template: source_sorting (Constrain by Source): corrected explanation to reflect that the behavior is actually the same in both modes, since only one match is ever marked as selected, and that match should always come first

bugfix: lib/sh/util.sh: str2varname(): need to lowercase str because on case-insensitive filesystems, paths sometimes canonicalize to a different capitalization than the original

lib/sh/util.sh: added lowercase()

bugfix: lib/sh/util.sh: die(): need stub since this is invoked before it's defined

bugfix: lib/sh/util.sh: setup_log_fd(): don't change $log_fd to stdlog until stdlog is set up, to avoid "Bad file descriptor" errors

lib/sh/util.sh: func_override(), copy_func(): added echo_func to facilitate debugging

bugfix: lib/sh/util.sh: stubs: log++ alias also needs to be moved to stub section

bugfix: lib/Firefox_bookmarks.reformat.csv: URLs: match only the uppercase tags used by Firefox, not any lowercase tags added by the user

fix: lib/Firefox_bookmarks.reformat.csv: page's self-description: updated comment to match regexp

bugfix: lib/Firefox_bookmarks.reformat.csv: page's self-description: updated "page's self-description: " prefix to remove

lib/sh/sync.sh: db_snapshot(): before backing up, trim bloated temp files (eg. from rolled back changes)

bugfix: lib/sql_io.py: put_table(): handle_MissingCastException(): when updating join_cols, don't add new entry for join_cols[out_col], only update existing one. this fixes #902 (import bug), and with #902 fixed, #887 (disk space leak) should no longer occur.

bugfix: lib/Firefox_bookmarks.reformat.csv: updated for new Firefox bookmarks format, which indents the <DD> tag

lib/tnrs.py: dirty: documented that this actually used to be on in the web app (see r9910, 2013-6-18), but does not appear to be needed (the source_sorting bug alluded to in r9910 is not fixed by enabling the dirty setting)

lib/tnrs.py: requests: also debug-print request URL

lib/tnrs.py: Download: include the same debug info as do_request()

lib/tnrs.py: do_request(): also debug-print request headers

lib/tnrs.py: download_request_template: dirty: documented why this must be off

bugfix: lib/tnrs.py: download_request_template: fixed bug where multiple names were being marked as Selected, because dirty was incorrectly set to true unlike in the web app

fix: lib/phpPgAdmin.login.php.diff: use relative file path rather than the path the file was at when the patch was created

/Makefile, lib/phpPgAdmin.login.php.diff: public_ user: added auto-filled password so that users would not be confused as to what to type in the password field

lib/tnrs.py: source_sorting (Constrain by Source): documented the different behavior for this in each match mode (all-matches and best-match)

fix: lib/runscripts/extract.run: export_(): explicitly prevent files from becoming web-accessible, to protect against an incorrect umask in the calling process

lib/tnrs.py: max_names: raised back up to 500 now that a workaround for the Internal Server Errors is in place (https://github.com/iPlantCollaborativeOpenSource/TNRS/issues/7)

fix: lib/tnrs.py: max_names: lowered to 50 because the dev TNRS server is now always crashing with an Internal Server Error when scrubbing 500 names at a time (https://github.com/iPlantCollaborativeOpenSource/TNRS/issues/7)

fix: lib/tnrs.py: Constrain by Source: turn it on so that the download settings reflect what TNRS actually used, while this is broken

fix: lib/tnrs.py: max_names: reduced back to 500 because even 5000 crashes the dev TNRS server

lib/tnrs.py: max_names: reduced to 5000 because 100,000 causes an internal server error

lib/tnrs.py: switched to downloading all matches per name, as is needed to implement #917. note that this will break the parts of the schema that use the tnrs table, until Brad's match-picking algorithm can be implemented, but this tradeoff is necessary to be able to begin scrubbing sooner (Martha; wiki.vegpath.org/2014-05-29_conference_call#TNRS)

lib/tnrs.py: max_names: increased to 100000 because the dev server can handle more names (no simultaneous users), as decided in the conference call (wiki.vegpath.org/2014-05-29_conference_call#TNRS)

fix: lib/PostgreSQL-MySQL.csv: need to replace "double precision" with "double" to work with MySQL Workbench 5.2.47

lib/tnrs.py: commented out the value of max_names that is not active, for clarity

lib/tnrs.py: sources: updated to list/sort order in issue #917

fix: lib/PostgreSQL-MySQL.csv: also remove left-behind lines such as `$$);`

bugfix: lib/runscripts/util.run: $is_runscript: unexport so don't pass it to invoked scripts

bugfix: lib/runscripts/util.run: run_args_cmd(): don't prepend main to args if no args, because for a non-runscript, all args will be passed to main(), leading `main` to be doubled

lib/tnrs.py: use the TNRS dev server (with private URL in tnrs.url) instead of the live server, since that contains datasources that we need

lib/streams.py: added file_get_contents()

lib/tnrs.py: configure the server separately from the base URL

lib/: svn:ignore tnrs.url so the TNRS dev server URL does not become public

lib/tnrs.py: retrieval_request_template: taxonomic_constraint, source_sorting: documented their meaning and why they need to be on/off

lib/runscripts/import.run: added install() target

lib/runscripts/in_datasrc_dir.run: use new local.run

added lib/runscripts/local.run

fix: lib/util.py: dict_subset(): raise an error if collections.OrderedDict isn't available, because some callers may depend on this. note that using dict instead of OrderedDict may be the cause of the joining on the wrong columns bug (issue #902).

bugfix: lib/runscripts/validations.pg.sql.run: updated to reflect that validations.sql is now located inside a subdir, not the datasrc dir

fix: lib/runscripts/file.pg.sql.run: removed include of in_datasrc_dir.run, because this location does not apply to all .sql export scripts

lib/runscripts/validations.pg.sql.run: export_(): make the export idempotent for easier re-runnability

bugfix: lib/sh/db.sh: pg_dump(): need use_pg to import $pg_database before checking for existence of $database

lib/sh/util.sh: import_vars: documented that it's idempotent

bugfix: lib/util.py: use OrderedDict from collections rather than ordereddict to work with Mac OS X 10.8 Mountain Lion (http://vegpath.org/links/#OrderedDict)

bugfix: benign_does_not_exist_error(): removed ignore_e=3, because this exit status is also used for other errors

fix: lib/sh/db.sh: benign_does_not_exist_error(): use benign_error=1, which is now supported properly by stderr_matches()

bugfix: lib/sh/util.sh: stderr_matches(): support $benign_error properly, by handling exit status logging in this func instead

bugfix: lib/sh/db.sh: pg_schema_exists(): also need to benignify "does not exist" error if returns false

bugfix: lib/sh/util.sh: stderr_matches(): need to separately display errors that were incorrectly suppressed due to $benign_error

bugfix: lib/sh/util.sh: is_err(): rethrow must be inverted (rethrow->false if error)

lib/sh/util.sh: added is_err()

lib/sh/local.sh: public_schema_exists(): moved to lib/sh/db.sh since this no longer depends on BIEN-specific configurations

bugfix: lib/sh/db.sh: public_schema_exists(): don't hide the function call tree so it's clear which function is running the psql commands

bugfix: lib/sh/db.sh: public_schema_exists(): don't hide the function call tree so it's clear which function is running the psql commands

fix: *{.sh,run}: stderr_matches() callers: added benign_error=1 where needed

fix: *{.sh,run}: stderr_matches() callers: usage: documented that they may require benign_error=1

fix: lib/sh/util.sh: stderr_matches(): usage: documented that this may require benign_error=1

bugfix: lib/sh/db.sh: pg_cmd(): updated for new echo_vars log_level

fix: lib/sh/db.sh: pg_schema_exists(): display the function name so it's clear which function is running the psql commands

fix: lib/sh/db.sh: pg_schema_exists(): don't use log++ because it hides the command that produces the benign error

fix: lib/sh/db.sh: psql(): removed debugging changes

bugfix: lib/sh/util.sh: highlight_log_msg(): when not can_highlight_log_msg, need to remove any surrounding formatting

fix: lib/sh/util.sh: die_error_hidden(): always log local vars at same log_level as echo_func

fix: *{.sh,run}: always log kw_params at same log_level as echo_func

lib/sh/util.sh: split_lines(): usage: matched up and synced different syntaxes

bugfix: lib/sh/util.sh: log_msg!(): split_lines does not support being invoked by wrapper; need to use `declare lines; wrapper "split_lines" str` instead

fix: lib/sh/util.sh: split_lines(): usage: documented different syntax for when using wrapper

bugfix: lib/sh/util.sh: die_error_hidden(): echo_func to assist debugging

bugfix: lib/sh/util.sh: split(): need to limit the effects of IFS to just the splitting, so it doesn't cause strange errors in other functions

bugfix: stderr2stdout(): fd 2 must be redirected back to fd 2, not log-filtered, in case there are other errors in addition to the benign error. this fixes a bug in pg_schema_exists(), where errors about the DB being down were not displayed because they were log-filtered out.

lib/sh/make.sh: set_make_vars: don't display make vars at verbosity 2 to avoid clutter

bugfix: lib/sh/make.sh: begin_target: don't echo_func twice

bugfix: lib/runscripts/util.run: a non-runscript should have all args passed to main(). this fixes a bug in backups/*_snapshot where "main" would need to be prepended to any args for the script to run correctly.

bugfix: lib/runscripts/util.run: $wrap_fn: invoked script must always run as runscript so that wrapped command is run

lib/runscripts/util.run: added $is_runscript, for use by $wrap_fn

bugfix: lib/runscripts/util.run: $wrap_fn: $top_script doesn't need to be world-executable for most uses of sudo (only if sudoing to non-root)

lib/sh/db.sh: pg_snapshot(): perform online backup if possible, to avoid needing to restart the server

lib/sh/db.sh: added pg_start_backup(), pg_stop_backup()

lib/sh/db.sh: psql(): only set redirs if can redirect

bugfix: psql(): when as_root is on, need to avoid redirections as these are not passed through by sudo

bugfix: lib/sh/db.sh: psql(): don't use `--file /dev/fd/40` when can't redirect

fix: lib/sh/db.sh: psql(): when using sudo with custom $stdin, raise error that this is not supported

lib/sh/db.sh: psql(): $data_bypasses_filter: renamed to $data2stdout for clarity

lib/sh/db.sh: psql(): $bypass_ok: renamed to $can_redir for clarity