streamlet-examples/pdftotext
author František Kučera <franta-hg@frantovo.cz>
Sat, 06 Jun 2020 01:50:42 +0200
branchv_0
changeset 80 fd7b3e510dd9
parent 50 22ed5647b235
permissions -rwxr-xr-x
Added tag v0.16 for changeset 288a7f075777
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
33
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     1
#!/bin/bash
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     2
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     3
# Relational pipes
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     4
# Copyright © 2020 František Kučera (Frantovo.cz, GlobalCode.info)
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     5
#
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     6
# This program is free software: you can redistribute it and/or modify
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     7
# it under the terms of the GNU General Public License as published by
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     8
# the Free Software Foundation, version 3 of the License.
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
     9
#
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    10
# This program is distributed in the hope that it will be useful,
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    11
# but WITHOUT ANY WARRANTY; without even the implied warranty of
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    12
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    13
# GNU General Public License for more details.
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    14
#
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    15
# You should have received a copy of the GNU General Public License
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    16
# along with this program. If not, see <http://www.gnu.org/licenses/>.
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    17
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    18
42
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    19
# This streamlet provides a single attribute: plain text content of given PDF file. It calls the tool pdftotext.
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    20
# n.b. the plain text content must fit into memory and shell variable and command-line argument (it usually will)
33
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    21
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    22
50
22ed5647b235 streamlets: include streamlet-common.sh from the directory where the actual streamlet resides (not where is the symlink, if any)
František Kučera <franta-hg@frantovo.cz>
parents: 49
diff changeset
    23
. "$(dirname "$(realpath "$0")")/streamlet-common.sh"
33
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    24
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    25
processMessage_WAITING_FOR_OUTPUT_ATTRIBUTES_METADATA() {
42
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    26
	send OUTPUT_ATTRIBUTE_METADATA "${outputAttributeAliases[0]-pdftotext}"    "string"
33
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    27
	send WAITING_FOR_INPUT_ATTRIBUTES
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    28
}
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    29
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    30
processMessage_WAITING_FOR_OUTPUT_ATTRIBUTES() {
42
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    31
	value="$(pdftotext "$currentFile" - | tr -d \\f)"; # tr just removes page breaks
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    32
	if   [[ "x$?" == "x0" ]]; then isNull="false"; else value=""; isNull="true"; fi
f1bbcf616269 streamlet examples: pdftotext
František Kučera <franta-hg@frantovo.cz>
parents: 33
diff changeset
    33
	send OUTPUT_ATTRIBUTE "$value"    "$isNull";
33
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    34
	send WAITING_FOR_INPUT_ATTRIBUTES;
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    35
}
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    36
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    37
initialize
f9cada1d46a4 streamlet examples: common functions + inode, lines_count, mime_type
František Kučera <franta-hg@frantovo.cz>
parents:
diff changeset
    38
processMessages