#!/usr/bin/env perl
use warnings;
use strict;
use 5.008003;
no warnings 'utf8';
our $VERSION = '2.006';
use Encode::Locale qw();
use App::DBBrowser;
use if $^O eq 'MSWin32', 'Win32::Console::ANSI';
print "\e(U" if $^O eq 'MSWin32';
binmode STDIN, ':encoding(console_in)';
binmode STDOUT, ':encoding(console_out)';
binmode STDERR, ':encoding(console_out)';
my $db_browser = App::DBBrowser->new();
$db_browser->run();
__END__
=pod
=encoding UTF-8
=head1 NAME
C<db-browser> - Browse C<SQLite>/C<MySQL>/C<PostgreSQL> databases and their tables interactively.
=head1 VERSION
Version 2.006
=cut
=head1 SYNOPSIS
=head2 SQLite/MySQL/PostgreSQL
db-browser -h|--help
db-browser
db-browser [database-name, ...]
When the C<db-browser> is called with the argument C<-h|--help>, it shows a menu. The menu entry I<HELP> shows this
documentation - see L</OPTIONS>.
If C<db-browser> is called without arguments, the user can choose from the databases offered by the database plugin. The
database is chosen automatically if only one database is available. With the C<SQLite> driver in use, the option
I<Search directories> tells the database plugin where to search for C<SQLite> databases (defaults to the home directory).
If C<db-browser> is called with arguments, the arguments are used as the available databases.
=head2 SQLite
db-browser [-s|--search]
C<db-browser> called with C<-s|--search> causes a new search for C<SQLite> databases instead of using the cached data.
=head1 DESCRIPTION
Before using C<db-browser>, make sure you backed up your databases.
Search and read in SQL databases: one can browse databases and their tables interactively.
The database plugins bundled with C<App::DBBrowser> provide support for the DBI drivers C<DBD::SQLite>, C<DBD::mysql>
and C<DBD::Pg>. See L<App::DBBrowser::DB> how to write a database plugin.
Before the output leading and trailing spaces are removed from the elements and spaces are squashed to a single
white-space.
The elements in a column are right-justified if one or more elements of that column do not look like a number, else they
are left-justified.
See L<Term::TablePrint|Term::TablePrint/DESCRIPTION> for more details.
=head3 Legacy encodings
Non mappable characters will break the output.
=head1 USAGE
The best way to find out how C<db-browser> works is calling C<db-browser>.
=head3 Keys to move around
=over
=item *
the C<Arrow> keys (or C<h,j,k,l>) to move up and down and to move to the right and to the left.
=item *
the C<PageUp> key (or C<Ctrl-B>) to go back one page, the C<PageDown> key (or C<Ctrl-F>) to go forward one page.
=item *
the C<Home> key (or C<Ctrl-A>) to jump to the beginning of the menu, the C<End> key (or C<Ctrl-E>) to jump to the end
of the menu.
=back
With the option I<mouse> enabled it can be used the mouse with the left mouse key to navigate through the menus.
To confirm a chosen menu item use the C<Return> key.
In some sub-menus it is possible to select more then one item before C<Return> is pressed; in such sub-menus the list of
items marked with the C<SpaceBar> key including the highlighted item are added to the chosen items when C<Return>
is pressed. If a mouse mode is enabled, it can be used the right mouse key instead of the C<SpaceBar>.
To leave a "readline" without entering anything enter the C<eof> control character (C<Ctrl-D> for *nix). Pressing only
the C<Enter> key means entering an empty string.
=head3 SQL menu
The SQL menu is the menu which opens after a table was selected.
If C<AGGREGATE> or C<GROUP BY> is set, the C<SELECT> statement is automatically formed; a previous user defined
C<SELECT> statement is reset. A user defined C<SELECT> resets a previous set C<AGGREGATE> or C<GROUP BY> statement.
To reset a SQL "sub-statement" (e.g C<WHERE>) re-enter into the respective menu and choose C<-OK->.
Changing the I<lock> mode (C<Lk0>,C<Lk1>) resets the entire SQL.
See option I<Expert> for the use of subqueries.
=head3 Delete, Update and Insert
To get to the I<Delete>/I<Update>/I<Insert> menu, navigate to the SQL menu and then select the prompt line.
Three different ways to insert data into a table are available:
=head4 plain
Insert data column by column.
=head4 copy and paste
Multi line input: reads until the end of input. It is OS-depend how to indicate the end of input.
In this mode leading and trailing whitespaces of the fields/columns are removed.
=head4 from file
Read the input from am file. Supported file formats: text-files and file formats supported by L<Spreadsheet::Read>.
Setting the insert mode to I<from file> or I<copy and paste> enables different filters: it possible to choose the
columns and rows which should be inserted.
C<DELETE>, C<UPDATE> or C<INSERT INTO> are not available with C<JOIN> and C<UNION> statements (except C<INSERT> with
C<mysql> and C<JOIN>).
=head3 Create table, Drop table
To reach the I<Create table>/I<Drop table> menu, select the prompt in the menu where the tables are selected.
I<Drop table>: the table is shown (here I<Auto Limit> has no effect, the whole table is fetched) before the user
confirms to drop the table.
For C<SQLite> databases there are additional entries are available in this menu: I<Attach DB> to attach databases to the
current database and if the current database has attached databases also I<Detach DB>.
=head3 Scalar functions
The scalar functions can be reached in the main SQL menu and also in the C<DELETE> and C<UPDATE> SQL sub-menus by
selecting the entry I<Func>.
The available functions are:
=head4 Epoch_to_DateTime
=head4 Epoch_to_Date
=head4 Truncate
With C<SQLite> the function C<TRUNCATE> is a user-defined function which returns stringified values.
return sprintf "%.*f", $places, int( $number * 10 ** $places ) / 10 ** $places;
When comparing in C<WHERE> or C<HAVING> clauses with numbers, take the non-truncated (original) value for the
comparison.
Also to get a numeric sort in an C<ORDER BY> clause use the non-truncated (original) values for the ordering.
=head4 Bit_Length
With C<SQLite> the function C<Bit_Length> is a user-defined function which uses the Perl builtin C<length>. To make
C<length> return the number of bytes the C<bytes> pragma is C<use>d.
=head4 Char_Length
With C<SQLite> the function C<Char_Length> is a user-defined function which uses the Perl builtin C<length> to get the
number of characters.
To reset a modified a column select the column with the function a second time.
=head2 OPTIONS
The options menu is called with C<db-browser -h>.
=head3 HELP
Show this Info.
=head3 Path
Shows the version and the path of the running C<db-browser> and the path of the application directory.
=head3 Database
=head4 DB Plugins
Choose the required database plugins.
=head4 DB Settings
These driver specific I<DB Settings> are used as the default database settings.
There is also in each database sub-menu the menu entry I<Database settings>. If these database specific parameter are not
set, the global (to the database plugin) I<DB Settings> are use instead.
User defined database plugins: it depends on the plugin which items are offered to set in each option and whether the
selections made by the user are considered.
=over
=item Fields
Set which fields are required to connect to a database.
=item ENV Variables
The user can choose environment variables form a list of environment variables that should be used to connect to the
database.
=item Login Data
The entered login data is saved in a configuration file and used to connect to the database (the password can not be
saved).
=item Attributes
For the meaning of these driver specific attributes check the appropriate driver documentation.
=item SQLite directories
This is a C<SQLite>-only option.
Sets the directories where C<db-browser> searches for C<SQLite> databases. Defaults to the home directory.
To move around in the directory tree select a directory and press C<Return> to enter in the selected directory or choose
C< .. > to move upwards. To add the current working-directory to the list of chosen directories use the C< . > (single
dot) menu entry. Select C< = > to confirm the made choices. The C< < > menu entry resets the list of chosen directories if
any. If the list of chosen directories is empty, C< < > goes back without changing anything.
This setting can not be overwritten in a single database.
=item Reset DB
Reset database specific parameter to the global I<DB Settings>.
=back
=head3 Menu
Set the behavior of the interactive menus.
=head4 Menu Memory
If enabled: saves the menu position while entering in a sub menu.
=head4 Table
Expand Rows:
=over
=item C<YES>
if C<Return> is pressed, the selected table row is printed with each column in its own line.
=item C<NO>
don't expand table rows.
=item C<YES fast back>
do not expand the first row if the cursor auto-jumped to the first row.
=back
=head4 Mouse Mode
Set the I<Mouse Mode> (see L<Term::Choose/mouse>).
=head3 SQL
=head4 Metadata
If I<Metadata> is enabled, system tables/schemas/databases are appended to the respective list.
=head4 Operators
Choose the required operators.
There are two regexp entries: C<REGEXP> matches case sensitive while C<REGEXP_i> matches case insensitive.
With C<MySQL> the sensitive match is achieved by enabling the C<BINARY> operator.
=head4 Lock
Set the default I<lock> value:
- C<Lk0>: Reset the SQL-statement after each time a table is printed.
- C<Lk1>: Reset the SQL-statement only when a table is selected.
=head4 Expert (experimental)
=over
=item I<Aggregate in SELECT>
Enables aggregate function in the SELECT substatement C<&>.
=item I<Subqueries>
If this option is enabled, the last 20 queries are saved temporary. Saved queries can than be chosen as subqueries by
selecting C<(Q>.
A query is saved when the table is printed. If a new database is selected the query-history is removed.
=item I<Alias for complex columns>
If enable, it is asked for an alias for columns with a function, for subqueries, for joined tables and for unions. If
entered nothing, no alias is added except for joined tables and unions which get a default alias.
=item I<Parentheses>
Enable parentheses in C<WHERE> and/or C<HAVING> clauses.
=back
=head4 Identifiers
- I<Qualified table names>: if set to C<YES> qualified table names are used in SQL statements.
- I<Quote identifiers>: if set to C<YES> SQL identifiers are quoted.
C<SQLite>:
- databases in SQL statements (C<ATTACH DATABASE>) are always quoted.
- if the current database has attached databases, the use of qualified table names is enabled automatically.
=head4 Write access
Enable write access - use with care.
=head3 Output
=head4 Max Rows
Sets C<LIMIT> automatically to I<Max Rows>. This can be overwritten by setting a SQL C<LIMIT> statement manually. Fetched
table rows are kept in memory.
To disable the automatic limit set I<Max Rows> to C<0>.
=head4 Colwidth
Columns with a width below or equal I<Colwidth> are only trimmed if it is still required to lower the row width despite
all columns wider than I<Colwidth> have been trimmed to I<Colwidth>.
=head4 ProgressBar
Set the progress bar threshold. If the number of fields (rows x columns) is higher than the threshold, a progress bar is
shown while preparing the data for the output.
=head4 Tabwidth
Set the number of spaces between columns.
=head4 Grid
Separate the columns from each other and the header from the body with lines.
=head4 Keep Header
Show the table header on top of each page.
=head4 Undef
Set the string that will be shown on the screen instead of an undefined field.
On C<MSWin32> only single-byte character sets are supported when entering the setting I<Undef>, I<user>, I<host> or I<port>
with the C<db-browser>s readline. But it is possible to edit the entry in configuration files directly after the entry
was created with this options menu.
=head4 Binary binary
Setting the I<binary_filter> to C<1> means: print "BNRY" instead of arbitrary binary data. If data matches the repexp
C</[\x00-\x08\x0B-\x0C\x0E-\x1F]/>, it is considered arbitrary binary data. Printing arbitrary binary data could break
the output.
=head3 Insert
These I<Insert> settings can also be set in the C<INSERT INTO> sub-statement menu.
=head4 File History
Set how many input file names should be saved. A value of C<0> disables the file history.
=head4 File encoding
How to decode text files.
=head4 Parse mode
Set how to parse text files or the "multi row" input. Files where C<-T $filename> returns true are considered text
files.
If a file is not a text file, then it is always used C<Spreadsheet::Read> to parse the file regardless of this
setting.
=over
=item
Text::CSV
To decode the files it is used the I<File encoding>.
Advantages:
=over
=item
Allows to set different csv-related options.
=back
=back
=over
=item
split
Reads to whole input at once and splits the input with the input record separator (IRS) to get the records (rows). Then
it splits the records with the input field separator (IFS) to get the fields (columns) of each record.
To decode the files it is used the I<File encoding>.
Advantages:
=over
=item
The values assigned to the IRS and the IFS are treated as regexps.
=back
=back
=over
=item
Spreadsheet::Read
If L<Spreadsheet::Read> is chosen, the default settings from C<Spreadsheet::Read> are used.
C<Spreadsheet::Read> will use the first line of the file to auto-detect the separation character if the file is a
csv-file.
Advantages:
=over
=item
Auto-detects the separation character for csv-files
=back
=back
=over
=back
=head4 Config Text::CSV
Set different L<Text::CSV> options.
=head4 Config 'split'
=over
=item
Record separator
Set the input record separator (regexp).
=item
Field separator
Set the input field separator (regexp).
=item
Trim leading
Expects a regex pattern. If set removes leading characters matching regexp from each field.
=item
Trim trailing
Expects a regex pattern. If set removes trailing characters matching regexp from each field.
=back
=head4 Create table
=over
=item
Default data type
The default data type of the columns.
=back
=head2 CONFIGURATION FILES
To find out the location of the configuration files call C<db-browser -h> and choose I<Path>. The data is saved in JSON
format.
=head1 REQUIREMENTS
See L<Term::TablePrint/REQUIREMENTS>.
=head2 Perl version
Requires Perl version 5.8.3 or greater.
=head2 Decoded strings
C<db-browser> expects decoded strings.
=head2 Terminal
It is required a terminal that uses a monospaced font which supports the printed characters.
The terminal has to understand ANSI escape sequences. If the OS is MSWin32 C<App::DBBrowser> uses
L<Win32::Console::ANSI> which emulates an ANSI console for the C<db-browser>.
=head2 Permissions
To be able to browse the database-, schema- and table-lists and the content of tables the user must have the database
privileges required for fetching the requested data.
The C<db-browser> expects an existing home directory with read and write permissions for the user of the C<db-browser>.
=head1 CREDITS
Thanks to the L<Perl-Community.de|http://www.perl-community.de> and the people form
L<stackoverflow|http://stackoverflow.com> for the help.
=head1 AUTHOR
Matthäus Kiem <cuer2s@gmail.com>
=head1 LICENSE AND COPYRIGHT
Copyright 2012-2018 Matthäus Kiem.
THIS SOFTWARE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE
IMPLIED WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl 5.10.0. For
details, see the full text of the licenses in the file LICENSE.
=cut