ETL from SFTP/CSV to MENTDB/TXT.mql
ETL script with application 'supervision' compatibility
if (not (group exist "0001_folder1_folder2_folder3")) {
group add "0001_folder1_folder2_folder3";
};
script merge "folder1.folder2.folder3.step_1_source_connect.exe" true 5
(param
)
"Connect to the source"
{
# CONFIGURATION ;
-> "[CONF_NAME_OF_THE_FLOW]" "name_of_the_flow";
-> "[CONF_SOURCE_CM]" "demo_cm";
# INITIALIZATION ;
-> "[FLOW_PID]" [PID];
# HANDLE ;
try {
# Flow initialization;
stack flow_init [FLOW_PID] [CONF_NAME_OF_THE_FLOW] "{}";
stack flow_step [FLOW_PID] 1 "source_connect...";
sftp connect "session_remote" {cm get [CONF_SOURCE_CM];};
#Step 1 is valid;
stack flow_step [FLOW_PID] 1 "source_connect_ok";
include "folder1.folder2.folder3.step_2_extract.exe";
} {
try {sftp disconnect "session_remote";} {} "[err]";
#Step 1 is not valid;
stack flow_step [FLOW_PID] 1 "source_connect_KO";
# Give the error to the stack and stop the process;
exception (1) ([global_err]);
} "[global_err]";
} "Return nothing";
if (not (group is granted script "folder1.folder2.folder3.step_1_source_connect.exe" "0001_folder1_folder2_folder3")) {
group grant script "folder1.folder2.folder3.step_1_source_connect.exe" "0001_folder1_folder2_folder3";
};
script merge "folder1.folder2.folder3.step_2_extract.exe" true 1
(param
)
"Extract files from the source"
{
# CONFIGURATION ;
-> "[CONF_DIR_SOURCE]" "tmp/source";
-> "[CONF_SOURCE_FILTER]" "*csv";
# HANDLE ;
try {
# Mark the flow as step 2;
stack flow_step [FLOW_PID] 2 "extract...";
# Create the local directory PID;
file mkdir (concat "home/" [FLOW_PID]);
# Get files into PID directory;
log write (concat "Get files from '" [CONF_DIR_SOURCE] "/" [CONF_SOURCE_FILTER] "'.") OK null null;
sftp cd "session_remote" [CONF_DIR_SOURCE];
json load "valid_files" (ftps ls "session_remote" [CONF_SOURCE_FILTER]);
json parse_array "valid_files" "/" "obj" {
-> "[filename]" (json select "obj" /name);
sftp get "session_remote" [filename] (concat "home/" [FLOW_PID] "/" [filename]);
log write (concat "Get file 'home/" [FLOW_PID] "/" [filename] "'.") OK null null;
};
# Delete source file;
json parse_array "valid_files" "/" "obj" {
-> "[filename]" (json select "obj" /name);
sftp rm "session_remote" [filename];
log write (concat "Remove source file '" [CONF_DIR_SOURCE] "/" [filename] "'.") OK null null;
};
} {
try {sftp disconnect "session_remote";} {} "[err]";
#Step 2 is not valid;
stack flow_step [FLOW_PID] 2 "extract_KO";
# Give the error to the stack and stop the process;
exception (1) ([global_err]);
} "[global_err]";
log write (concat (json count "valid_files" /) " file(s) to transform.") OK null null;
if (> (json count "valid_files" /) 0) {
#Step 2 is valid;
stack flow_step [FLOW_PID] 2 (concat "extract_ok_" (json count "valid_files" /) "_files");
json parse_array "valid_files" "/" "obj" {
-> "[filename]" (json select "obj" /name);
stack (date now) "folder1.folder2.folder3.step_3_transform.exe" "[FLOW_PID]" [FLOW_PID] "[filename]" [filename];
};
} {
#Step 2 is valid;
stack flow_step [FLOW_PID] 2 "extract_ok_zero_file";
# No file into the directory;
file delete (concat "home/" [FLOW_PID]);
};
} "Return nothing";
if (not (group is granted script "folder1.folder2.folder3.step_2_extract.exe" "0001_folder1_folder2_folder3")) {
group grant script "folder1.folder2.folder3.step_2_extract.exe" "0001_folder1_folder2_folder3";
};
script merge "folder1.folder2.folder3.step_3_transform.exe" true 1
(param
(var "[FLOW_PID]" {true} "The flow PID" is_null:true is_empty:true "1")
(var "[filename]" {true} "The filename" is_null:true is_empty:true "file.xxx")
)
"Transform the file"
{
# CONFIGURATION ;
-> "[CONF_SOURCE_CSV_SEPARATOR]" ",";
-> "[CONF_SOURCE_CSV_QUOTE]" "'";
-> "[CONF_SOURCE_CSV_FORCE_COLUMN]" "A,B,C";
-> "[CONF_DESTINATION_NEW_FILENAME]" (concat "new_filename_" (date systimestamp_min) ".txt");
-> "[CONF_DESTINATION_TXT_SEPARATOR]" "|";
# HANDLE ;
try {
# Mark the flow as step 3;
stack flow_step [FLOW_PID] 3 "transform...";
file writer_open "w1" (concat "home/" [FLOW_PID] "/" [CONF_DESTINATION_NEW_FILENAME]) true TEXT "utf-8";
file writer_add_line "w1" (concat "HEADER");
file writer_flush "w1";
log write (concat "Build the file 'home/" [FLOW_PID] "/" [CONF_DESTINATION_NEW_FILENAME] "' from 'home/" [FLOW_PID] "/" [filename] "'...") OK null null;
csv parse (mql "T") (concat "home/" [FLOW_PID] "/" [filename]) (mql [CONF_SOURCE_CSV_SEPARATOR]) (mql [CONF_SOURCE_CSV_QUOTE]) (mql [CONF_SOURCE_CSV_FORCE_COLUMN]) {
-> "[line]" "";
concat_var "[line]" [_n_];
concat_var "[line]" (string lpad [T_A] "0" 10) [CONF_DESTINATION_TXT_SEPARATOR];
concat_var "[line]" (string lpad [T_B] "0" 10) [CONF_DESTINATION_TXT_SEPARATOR];
concat_var "[line]" (string lpad [T_C] "0" 10);
file writer_add_line "w1" [line];
file writer_flush "w1";
};
file writer_close "w1";
log write (concat "Builded.") OK null null;
#Step 3 is valid;
stack flow_step [FLOW_PID] 3 "transform_ok";
stack (date now) "folder1.folder2.folder3.step_4_destination_connect.exe" "[FLOW_PID]" [FLOW_PID] "[filename]" [CONF_DESTINATION_NEW_FILENAME];
} {
try {file writer_close "w1";} {} "[err]";
#Step 3 is not valid;
stack flow_step [FLOW_PID] 3 "transform_KO";
# Give the error to the stack and stop the process;
exception (1) ([global_err]);
} "[global_err]";
} "Return nothing";
if (not (group is granted script "folder1.folder2.folder3.step_3_transform.exe" "0001_folder1_folder2_folder3")) {
group grant script "folder1.folder2.folder3.step_3_transform.exe" "0001_folder1_folder2_folder3";
};
script merge "folder1.folder2.folder3.step_4_destination_connect.exe" true 5
(param
(var "[FLOW_PID]" {true} "The flow PID" is_null:true is_empty:true "1")
(var "[filename]" {true} "The filename" is_null:true is_empty:true "file.xxx")
)
"Connect to the destination"
{
# CONFIGURATION ;
-> "[CONF_DESTINATION_CM]" "demo_cm";
# HANDLE ;
try {
# Flow initialization;
stack flow_step [FLOW_PID] 4 "destination_connect...";
tunnel connect "destination_sess" {cm get [CONF_DESTINATION_CM];};
#Step 4 is valid;
stack flow_step [FLOW_PID] 4 "destination_connect_ok";
include "folder1.folder2.folder3.step_5_load.exe";
} {
try {tunnel disconnect "destination_sess";} {} "[err]";
#Step 4 is not valid;
stack flow_step [FLOW_PID] 4 "destination_connect_KO";
# Give the error to the stack and stop the process;
exception (1) ([global_err]);
} "[global_err]";
} "Return nothing";
if (not (group is granted script "folder1.folder2.folder3.step_4_destination_connect.exe" "0001_folder1_folder2_folder3")) {
group grant script "folder1.folder2.folder3.step_4_destination_connect.exe" "0001_folder1_folder2_folder3";
};
script merge "folder1.folder2.folder3.step_5_load.exe" true 1
(param
)
"Load file to the destination"
{
# CONFIGURATION ;
-> "[CONF_DIR_DESTINATION]" "tmp/destination";
# HANDLE ;
try {
# Mark the flow as step 5;
stack flow_step [FLOW_PID] 5 "load...";
log write (concat "Send '" (concat "home/" [FLOW_PID] "/" [filename]) "' to the '" (concat [CONF_DIR_DESTINATION] "/" [filename]) "' ...") OK null null;
# Copy file into the destination directory;
#Open a reader;
file reader_open "r1" (concat "home/" [FLOW_PID] "/" [filename]) BINARY null;
tunnel execute "destination_sess" (concat
"-> \"[remote_destination]\" \"" (mql encode (concat [CONF_DIR_DESTINATION] "/" [filename])) "\";"
(mql {
file writer_open "w1" [remote_destination] true BINARY null;
})
);
#Parse the file;
while (is not null (-> "[bytes]" (file reader_get_bytes "r1" 402400));) {
tunnel execute "destination_sess" (concat
"-> \"[bytes]\" \"" (mql encode [bytes]) "\";"
(mql {
file writer_add_bytes "w1" [bytes];
file writer_flush "w1";
})
);
};
tunnel execute "destination_sess" (concat
(mql {
#Force to write;
file writer_flush "w1";
file writer_close "w1";
})
);
#Close the reader;
file reader_close "r1";
log write "Sended." OK null null;
#Step 5 is valid;
stack flow_step [FLOW_PID] 5 "load_ok";
try {tunnel disconnect "destination_sess";} {} "[err]";
} {
#Close objects;
try {file reader_close "r1";} {} "[sub_err]";
try {tunnel execute "destination_sess" (concat
(mql {
try {file writer_close "w1";} {} "[err]";
})
);} {} "[err]";
try {tunnel disconnect "destination_sess";} {} "[err]";
#Step 5 is not valid;
stack flow_step [FLOW_PID] 5 "load_KO";
# Give the error to the stack and stop the process;
exception (1) ([global_err]);
} "[global_err]";
} "Return nothing";
if (not (group is granted script "folder1.folder2.folder3.step_5_load.exe" "0001_folder1_folder2_folder3")) {
group grant script "folder1.folder2.folder3.step_5_load.exe" "0001_folder1_folder2_folder3";
};