Gro-Tsen/gettweet.pl

## gettweet.pl
#! /usr/local/bin/perl -w

# gettweet.pl: Retrieve the content of tweets by id.
# Usage: getweet.pl -i <comma-separated-list>
# saves each in a file called <fulldate>-<id>.txt

# Requires API keys to be stored in ~/.twitterkeys (or $TWITTERKEYS)
# syntax being four lines starting "API-Key: ", "API-Key-Secret: ",
# "Access-Token: " and "Access-Token-Secret: " each followed by their value.

use strict;
use warnings;

use Getopt::Std;

use Net::Twitter;
use DateTime::Format::Strptime;
use Data::Dumper;

binmode STDERR, ":utf8";
binmode STDOUT, ":utf8";

# Get options
my %opts;
getopts("i:", \%opts);
# -i <ids>: retrieve specific tweets (comma separated list)

my $lookuplist;
if ( defined($opts{i}) ) {
    die "Badly formed argument to -i (expecting a comma-separated list of integers)" unless $opts{i} =~ m/^[0-9]+(?:\,[0-9]+)*$/;
    $lookuplist = [ split(/,/, $opts{i}) ];
} else {
    die "Need to pass -i option";
}

# Read authentication file
my $keyfname = $ENV{"TWITTERKEYS"} // ($ENV{"HOME"}."/.twitterkeys");
my %keyfdata;
open my $keyf, "<", $keyfname or die "Can't open $keyfname: $!";
while (<$keyf>) {
    s/\#.*//;
    next if m/^\s+$/;
    die "Bad key file format: $_" unless m/^\s*([A-Za-z][A-Za-z0-9\-]*)\:\s+(\S+)\s*$/;
    # print STDERR "Read auth data \[$1\]=$2\n";
    $keyfdata{$1} = $2;
}
close $keyf;
my $consumer_key = $keyfdata{"API-Key"} or die "Didn't find API-Key in key file";
my $consumer_secret = $keyfdata{"API-Key-Secret"} or die "Didn't find API-Key-Secret in key file";
my $access_token = $keyfdata{"Access-Token"} or die "Didn't find Access-Token in key file";
my $access_token_secret = $keyfdata{"Access-Token-Secret"} or die "Didn't find Access-Token-Secret in key file";
die "Badly formed Access-Token" unless $access_token =~ m/^([0-9]+)\-/;
my $self_id = $1;

# Open the Twitter connection.
my $nt = Net::Twitter->new(
    traits => [qw/API::RESTv1_1/],
    consumer_key => $consumer_key,
    consumer_secret => $consumer_secret,
    access_token => $access_token,
    access_token_secret => $access_token_secret,
);

die "Couldn't open Twitter connection" unless $nt;

my $datetime_parser = DateTime::Format::Strptime->new(
    pattern => "%a %b %d %T %z %Y", time_zone => "UTC", locale => "C",
    on_error => "croak");

# Get self screen name
print STDERR "Getting information on self...\n";
my $slf = $nt->lookup_users({user_id => $self_id});
die "lookup_user didn't return exactly one user" unless scalar(@{$slf})==1;
die "Missing screen name" unless defined($slf->[0]->{"screen_name"});
my $self_screen_name = $slf->[0]->{"screen_name"};
my $self_name = $slf->[0]->{"name"};

my %paramdict;
my $tweetlist;
if ( 1 ) {
    %paramdict = (trim_user => 0, tweet_mode => "extended");
    $paramdict{id} = $lookuplist;
    $tweetlist = $nt->lookup_statuses(\%paramdict);
}

# The following function performs a set of substitutions on a string
# (correctly adjusting the position of each subsequent substitution to
# take into account earlier ones).
sub substitute_in_string {
    my $str = shift;
    my @subs = sort { $a->[0] <=> $b->[0] } @{ shift() };
    # Each entry of @subs is an array reference:
    # [0]: first character of substring to modify
    # [1]: length of substring to modify
    # [2]: undefined, or substring itself, to check for consistency
    # [3]: undefined, or substring to substitute
    # [4]: undefined, or substring to insert before
    # [5]: undefined, or substring to insert at end
    my $corr = 0;
    my $minbar = 0;
    for my $sb ( @subs ) {
	my $idx0 = $sb->[0] + $corr;
	die "Attempting to overlap substitutions" unless $idx0>=$minbar;
	my $len = $sb->[1];
	if ( defined($sb->[2]) ) {
	    die(sprintf("Bad calibration: expecting \"%s\", got \"%s\"\n", $sb->[2], substr($str, $idx0, $len))) unless substr($str, $idx0, $len) eq $sb->[2];
	}
	my $repl = ($sb->[4] // "") . ($sb->[3] // substr($str, $idx0, $len)) . ($sb->[5] // "");
	my $newlen = length($repl);
	# printf STDERR "str: %s\nsubstituting (%d,%d) \"%s\"->\"%s\"\n", $str, $idx0, $len, substr($str, $idx0, $len), $repl;
	substr($str, $idx0, $len) = $repl;
	# printf STDERR "str: %s\n", $str;
	$corr += ($newlen - $len);
	$minbar = $idx0 + $newlen;
    }
    return $str;
}

# Now scan through tweetlist
for my $st ( @{$tweetlist} ) {
    die "Missing id from tweetlist entry" unless defined($st->{"id"});
    die "Bad id from tweetlist entry" unless $st->{"id"} =~ /^[0-9]+$/;
    my $created_at = $datetime_parser->parse_datetime($st->{"created_at"});
    my $timestamp = $created_at->strftime("%Y%m%d%H%M%S");
    my $id = $st->{"id"};
    printf STDERR "%s-%s\n", $timestamp, $id;
    my $fname = $timestamp . "-" . $id . ".txt";
    my $user_id = $st->{"user"}->{"id"};
    my $user_screen_name = $user_id eq $self_id ? $self_screen_name : $st->{"user"}->{"screen_name"};
    die "Didn't get user screen name" unless defined($user_screen_name);
    my $permalink = sprintf("https://twitter.com/%s/status/%s", $user_screen_name, $id);
    my $fulltext = $st->{"full_text"};
    die "Text contains unescaped HTML" if $fulltext =~ m/[\<\>]|\&(?!(?:lt|gt|amp|apos|quot|\#(?:x[0-9A-Fa-f]+|[0-9]+))\;)/;
    my @substitutions = ();
    for my $ent ( @{$st->{"entities"}->{"urls"}} ) {
	my $idx0 = $ent->{"indices"}->[0];
	my $idx1 = $ent->{"indices"}->[1];
	push @substitutions, [$idx0, $idx1-$idx0, $ent->{"url"}, $ent->{"expanded_url"}];
    }
    for my $ent ( @{$st->{"entities"}->{"media"}} ) {
	my $idx0 = $ent->{"indices"}->[0];
	my $idx1 = $ent->{"indices"}->[1];
	push @substitutions, [$idx0, $idx1-$idx0, $ent->{"url"}, $ent->{"display_url"}];
    }
    $fulltext = substitute_in_string $fulltext, \@substitutions;
    # Save text archive in a file
    open my $f, ">:utf8", $fname or die "Cannot open $fname";
    print $f "$timestamp\n$id ($user_id)\n$permalink\n\n$fulltext\n";
    close $f;
}

## tweet.pl
#! /usr/local/bin/perl -w

# tweet.pl: Post a tweet on Twitter.
# Reads tweet content from stdin unless -c option is specified (giving content).
# Use -r id to reply to specified tweet ID.  Use of -m is preferable (required?).
# Use -n to chomp final NL (useful when reading from stdin).

# The id of the posted tweet is written on stdout, various diagnostics on stderr.

# Requires API keys to be stored in ~/.twitterkeys (or $TWITTERKEYS)
# syntax being four lines starting "API-Key: ", "API-Key-Secret: ",
# "Access-Token: " and "Access-Token-Secret: " each followed by their value.

use strict;
use warnings;

use Encode;

use Getopt::Std;

use Net::Twitter;
# use DateTime::Format::Strptime;
# use Data::Dumper;

use open IN => ':utf8';
binmode STDIN, ":utf8";
binmode STDERR, ":utf8";
binmode STDOUT, ":utf8";

# Get options
my %opts;
getopts("r:c:mn", \%opts);
# -r <id>: reply to specified ID
# -c <string>: status content
# -m: set auto_populate_reply_metadata
# -n: chomp final NL

my $in_reply_to;
if ( defined($opts{r}) ) {
    die "Badly formed argument to -r" unless $opts{r} =~ m/^(?:https?\:\/\/twitter\.com\/[^\/]*\/status\/)?([0-9]+)$/;
    $in_reply_to = $1;
}

my $auto_populate_reply_metadata;
$auto_populate_reply_metadata = 1 if $opts{m};

# Read authentication file
my $keyfname = $ENV{"TWITTERKEYS"} // ($ENV{"HOME"}."/.twitterkeys");
my %keyfdata;
open my $keyf, "<", $keyfname or die "Can't open $keyfname: $!";
while (<$keyf>) {
    s/\#.*//;
    next if m/^\s+$/;
    die "Bad key file format: $_" unless m/^\s*([A-Za-z][A-Za-z0-9\-]*)\:\s+(\S+)\s*$/;
    # print STDERR "Read auth data \[$1\]=$2\n";
    $keyfdata{$1} = $2;
}
close $keyf;
my $consumer_key = $keyfdata{"API-Key"} or die "Didn't find API-Key in key file";
my $consumer_secret = $keyfdata{"API-Key-Secret"} or die "Didn't find API-Key-Secret in key file";
my $access_token = $keyfdata{"Access-Token"} or die "Didn't find Access-Token in key file";
my $access_token_secret = $keyfdata{"Access-Token-Secret"} or die "Didn't find Access-Token-Secret in key file";
die "Badly formed Access-Token" unless $access_token =~ m/^([0-9]+)\-/;
my $self_id = $1;

# Open the Twitter connection.
my $nt = Net::Twitter->new(
    traits => [qw/API::RESTv1_1/],
    consumer_key => $consumer_key,
    consumer_secret => $consumer_secret,
    access_token => $access_token,
    access_token_secret => $access_token_secret,
);

die "Couldn't open Twitter connection" unless $nt;

# my $datetime_parser = DateTime::Format::Strptime->new(
#     pattern => "%a %b %d %T %z %Y", time_zone => "UTC", locale => "C",
#     on_error => "croak");

# Read status to be tweeted
my $status;
if ( defined($opts{c}) ) {
    $status = decode("utf-8", $opts{c});
} else {
    print STDERR "Reading tweet content...\n";
    { local $/ = undef; $status = <>; }
}

chomp $status if defined($opts{n});

print STDERR "Replying to $in_reply_to\n" if $in_reply_to;
print STDERR "Tweet to be posted follows:\n$status\n";

my %hash;
$hash{status} = $status;
$hash{in_reply_to_status_id} = $in_reply_to if defined($in_reply_to);
$hash{auto_populate_reply_metadata} = 1 if $auto_populate_reply_metadata;
# Not sure if this is useful/desirable?
# <URL: https://developer.twitter.com/en/docs/tweets/tweet-updates.html >
$hash{tweet_mode} = "extended";
my $tweet = $nt->update(\%hash);
print STDERR "Posted\n";
print $tweet->{id} . "\n";
	#! /usr/local/bin/perl -w

	# gettweet.pl: Retrieve the content of tweets by id.
	# Usage: getweet.pl -i <comma-separated-list>
	# saves each in a file called <fulldate>-<id>.txt

	# Requires API keys to be stored in ~/.twitterkeys (or $TWITTERKEYS)
	# syntax being four lines starting "API-Key: ", "API-Key-Secret: ",
	# "Access-Token: " and "Access-Token-Secret: " each followed by their value.

	use strict;
	use warnings;

	use Getopt::Std;

	use Net::Twitter;
	use DateTime::Format::Strptime;
	use Data::Dumper;

	binmode STDERR, ":utf8";
	binmode STDOUT, ":utf8";

	# Get options
	my %opts;
	getopts("i:", \%opts);
	# -i <ids>: retrieve specific tweets (comma separated list)

	my $lookuplist;
	if ( defined($opts{i}) ) {
	die "Badly formed argument to -i (expecting a comma-separated list of integers)" unless $opts{i} =~ m/^[0-9]+(?:\,[0-9]+)*$/;
	$lookuplist = [ split(/,/, $opts{i}) ];
	} else {
	die "Need to pass -i option";
	}

	# Read authentication file
	my $keyfname = $ENV{"TWITTERKEYS"} // ($ENV{"HOME"}."/.twitterkeys");
	my %keyfdata;
	open my $keyf, "<", $keyfname or die "Can't open $keyfname: $!";
	while (<$keyf>) {
	s/\#.*//;
	next if m/^\s+$/;
	die "Bad key file format: $_" unless m/^\s([A-Za-z][A-Za-z0-9\-])\:\s+(\S+)\s*$/;
	# print STDERR "Read auth data \[$1\]=$2\n";
	$keyfdata{$1} = $2;
	}
	close $keyf;
	my $consumer_key = $keyfdata{"API-Key"} or die "Didn't find API-Key in key file";
	my $consumer_secret = $keyfdata{"API-Key-Secret"} or die "Didn't find API-Key-Secret in key file";
	my $access_token = $keyfdata{"Access-Token"} or die "Didn't find Access-Token in key file";
	my $access_token_secret = $keyfdata{"Access-Token-Secret"} or die "Didn't find Access-Token-Secret in key file";
	die "Badly formed Access-Token" unless $access_token =~ m/^([0-9]+)\-/;
	my $self_id = $1;

	# Open the Twitter connection.
	my $nt = Net::Twitter->new(
	traits => [qw/API::RESTv1_1/],
	consumer_key => $consumer_key,
	consumer_secret => $consumer_secret,
	access_token => $access_token,
	access_token_secret => $access_token_secret,
	);

	die "Couldn't open Twitter connection" unless $nt;

	my $datetime_parser = DateTime::Format::Strptime->new(
	pattern => "%a %b %d %T %z %Y", time_zone => "UTC", locale => "C",
	on_error => "croak");

	# Get self screen name
	print STDERR "Getting information on self...\n";
	my $slf = $nt->lookup_users({user_id => $self_id});
	die "lookup_user didn't return exactly one user" unless scalar(@{$slf})==1;
	die "Missing screen name" unless defined($slf->[0]->{"screen_name"});
	my $self_screen_name = $slf->[0]->{"screen_name"};
	my $self_name = $slf->[0]->{"name"};

	my %paramdict;
	my $tweetlist;
	if ( 1 ) {
	%paramdict = (trim_user => 0, tweet_mode => "extended");
	$paramdict{id} = $lookuplist;
	$tweetlist = $nt->lookup_statuses(\%paramdict);
	}

	# The following function performs a set of substitutions on a string
	# (correctly adjusting the position of each subsequent substitution to
	# take into account earlier ones).
	sub substitute_in_string {
	my $str = shift;
	my @subs = sort { $a->[0] <=> $b->[0] } @{ shift() };
	# Each entry of @subs is an array reference:
	# [0]: first character of substring to modify
	# [1]: length of substring to modify
	# [2]: undefined, or substring itself, to check for consistency
	# [3]: undefined, or substring to substitute
	# [4]: undefined, or substring to insert before
	# [5]: undefined, or substring to insert at end
	my $corr = 0;
	my $minbar = 0;
	for my $sb ( @subs ) {
	my $idx0 = $sb->[0] + $corr;
	die "Attempting to overlap substitutions" unless $idx0>=$minbar;
	my $len = $sb->[1];
	if ( defined($sb->[2]) ) {
	die(sprintf("Bad calibration: expecting \"%s\", got \"%s\"\n", $sb->[2], substr($str, $idx0, $len))) unless substr($str, $idx0, $len) eq $sb->[2];
	}
	my $repl = ($sb->[4] // "") . ($sb->[3] // substr($str, $idx0, $len)) . ($sb->[5] // "");
	my $newlen = length($repl);
	# printf STDERR "str: %s\nsubstituting (%d,%d) \"%s\"->\"%s\"\n", $str, $idx0, $len, substr($str, $idx0, $len), $repl;
	substr($str, $idx0, $len) = $repl;
	# printf STDERR "str: %s\n", $str;
	$corr += ($newlen - $len);
	$minbar = $idx0 + $newlen;
	}
	return $str;
	}

	# Now scan through tweetlist
	for my $st ( @{$tweetlist} ) {
	die "Missing id from tweetlist entry" unless defined($st->{"id"});
	die "Bad id from tweetlist entry" unless $st->{"id"} =~ /^[0-9]+$/;
	my $created_at = $datetime_parser->parse_datetime($st->{"created_at"});
	my $timestamp = $created_at->strftime("%Y%m%d%H%M%S");
	my $id = $st->{"id"};
	printf STDERR "%s-%s\n", $timestamp, $id;
	my $fname = $timestamp . "-" . $id . ".txt";
	my $user_id = $st->{"user"}->{"id"};
	my $user_screen_name = $user_id eq $self_id ? $self_screen_name : $st->{"user"}->{"screen_name"};
	die "Didn't get user screen name" unless defined($user_screen_name);
	my $permalink = sprintf("https://twitter.com/%s/status/%s", $user_screen_name, $id);
	my $fulltext = $st->{"full_text"};
	die "Text contains unescaped HTML" if $fulltext =~ m/[\<\>]\|\&(?!(?:lt\|gt\|amp\|apos\|quot\|\#(?:x[0-9A-Fa-f]+\|[0-9]+))\;)/;
	my @substitutions = ();
	for my $ent ( @{$st->{"entities"}->{"urls"}} ) {
	my $idx0 = $ent->{"indices"}->[0];
	my $idx1 = $ent->{"indices"}->[1];
	push @substitutions, [$idx0, $idx1-$idx0, $ent->{"url"}, $ent->{"expanded_url"}];
	}
	for my $ent ( @{$st->{"entities"}->{"media"}} ) {
	my $idx0 = $ent->{"indices"}->[0];
	my $idx1 = $ent->{"indices"}->[1];
	push @substitutions, [$idx0, $idx1-$idx0, $ent->{"url"}, $ent->{"display_url"}];
	}
	$fulltext = substitute_in_string $fulltext, \@substitutions;
	# Save text archive in a file
	open my $f, ">:utf8", $fname or die "Cannot open $fname";
	print $f "$timestamp\n$id ($user_id)\n$permalink\n\n$fulltext\n";
	close $f;
	}
	#! /usr/local/bin/perl -w

	# tweet.pl: Post a tweet on Twitter.
	# Reads tweet content from stdin unless -c option is specified (giving content).
	# Use -r id to reply to specified tweet ID. Use of -m is preferable (required?).
	# Use -n to chomp final NL (useful when reading from stdin).

	# The id of the posted tweet is written on stdout, various diagnostics on stderr.

	# Requires API keys to be stored in ~/.twitterkeys (or $TWITTERKEYS)
	# syntax being four lines starting "API-Key: ", "API-Key-Secret: ",
	# "Access-Token: " and "Access-Token-Secret: " each followed by their value.

	use strict;
	use warnings;

	use Encode;

	use Getopt::Std;

	use Net::Twitter;
	# use DateTime::Format::Strptime;
	# use Data::Dumper;

	use open IN => ':utf8';
	binmode STDIN, ":utf8";
	binmode STDERR, ":utf8";
	binmode STDOUT, ":utf8";

	# Get options
	my %opts;
	getopts("r:c:mn", \%opts);
	# -r <id>: reply to specified ID
	# -c <string>: status content
	# -m: set auto_populate_reply_metadata
	# -n: chomp final NL

	my $in_reply_to;
	if ( defined($opts{r}) ) {
	die "Badly formed argument to -r" unless $opts{r} =~ m/^(?:https?\:\/\/twitter\.com\/[^\/]*\/status\/)?([0-9]+)$/;
	$in_reply_to = $1;
	}

	my $auto_populate_reply_metadata;
	$auto_populate_reply_metadata = 1 if $opts{m};

	# Read authentication file
	my $keyfname = $ENV{"TWITTERKEYS"} // ($ENV{"HOME"}."/.twitterkeys");
	my %keyfdata;
	open my $keyf, "<", $keyfname or die "Can't open $keyfname: $!";
	while (<$keyf>) {
	s/\#.*//;
	next if m/^\s+$/;
	die "Bad key file format: $_" unless m/^\s([A-Za-z][A-Za-z0-9\-])\:\s+(\S+)\s*$/;
	# print STDERR "Read auth data \[$1\]=$2\n";
	$keyfdata{$1} = $2;
	}
	close $keyf;
	my $consumer_key = $keyfdata{"API-Key"} or die "Didn't find API-Key in key file";
	my $consumer_secret = $keyfdata{"API-Key-Secret"} or die "Didn't find API-Key-Secret in key file";
	my $access_token = $keyfdata{"Access-Token"} or die "Didn't find Access-Token in key file";
	my $access_token_secret = $keyfdata{"Access-Token-Secret"} or die "Didn't find Access-Token-Secret in key file";
	die "Badly formed Access-Token" unless $access_token =~ m/^([0-9]+)\-/;
	my $self_id = $1;

	# Open the Twitter connection.
	my $nt = Net::Twitter->new(
	traits => [qw/API::RESTv1_1/],
	consumer_key => $consumer_key,
	consumer_secret => $consumer_secret,
	access_token => $access_token,
	access_token_secret => $access_token_secret,
	);

	die "Couldn't open Twitter connection" unless $nt;

	# my $datetime_parser = DateTime::Format::Strptime->new(
	# pattern => "%a %b %d %T %z %Y", time_zone => "UTC", locale => "C",
	# on_error => "croak");

	# Read status to be tweeted
	my $status;
	if ( defined($opts{c}) ) {
	$status = decode("utf-8", $opts{c});
	} else {
	print STDERR "Reading tweet content...\n";
	{ local $/ = undef; $status = <>; }
	}

	chomp $status if defined($opts{n});

	print STDERR "Replying to $in_reply_to\n" if $in_reply_to;
	print STDERR "Tweet to be posted follows:\n$status\n";

	my %hash;
	$hash{status} = $status;
	$hash{in_reply_to_status_id} = $in_reply_to if defined($in_reply_to);
	$hash{auto_populate_reply_metadata} = 1 if $auto_populate_reply_metadata;
	# Not sure if this is useful/desirable?
	# <URL: https://developer.twitter.com/en/docs/tweets/tweet-updates.html >
	$hash{tweet_mode} = "extended";
	my $tweet = $nt->update(\%hash);
	print STDERR "Posted\n";
	print $tweet->{id} . "\n";