\n';
'
\n';
); # if export_to_file
# ----------
# Set filter
# ----------
set_filter_expression(report_name, report_element);
# -----------------------------------------
# Query the database for the report-element
# -----------------------------------------
if (node_exists("volatile.query_result")) then (
delete_node("volatile.query_result");
);
# debug_message("report element node: " . report_element_node . "\n");
query_db_for_view(report_element_node, "volatile.query_result");
string query_header = ("volatile.query_result.header");
string query_data = ("volatile.query_result.data");
# debug_message("export query header: \n" . node_as_string(query_header) . "\n\n");
# debug_message("export query data: \n" . node_as_string(query_data) . "\n\n");
# =================================
# Handle aggregation_method average
# and total row sum setup
# =================================
# If there is a column where aggregation_method is average then calculate
# the average now so that the table becomes properly sorted.
# We add the sum for the total row in the query_header
# be adding a sum node
node column;
string field_name;
string data_type;
bool aggregation_method_is_sum;
bool aggregation_method_is_average;
node column_sum_node;
int count = 0;
bool handle_aggregation = false;
string database_fields_path = "profiles." . profile_name . ".database.fields";
if (node_exists("volatile.aggregation_columns_header")) then (
delete_node("volatile.aggregation_columns_header");
);
foreach column query_header (
data_type = node_value(subnode_by_name(column, "data_type"));
if (data_type ne "string") then (
field_name = node_value(subnode_by_name(column, "field_name"));
# ----------------------------
# handle sum (total row setup)
# ----------------------------
if (export_total_row) then (
aggregation_method_is_sum = aggr_method_is_sum(field_name, profile_name);
if (aggregation_method_is_sum and (data_type ne "unique")) then (
# prepare node to calculate the sum value
# in case of missing or invalid data type define float
if (data_type ne "int" and data_type ne "float") then (
data_type = "float";
);
column . ".sum" = 0;
column_sum_node = column . ".sum";
set_node_type(column_sum_node, data_type);
set_node_value(column_sum_node, 0);
);
);
# ---------------------------------
# handle aggregation method average
# ---------------------------------
aggregation_method_is_average = aggr_method_is_average(field_name, profile_name);
if (aggregation_method_is_average) then (
handle_aggregation = true;
"volatile.aggregation_columns_header." . count . ".field_name" = field_name;
"volatile.aggregation_columns_header." . count . ".average_denominator_field" = node_value(database_fields_path . "." . field_name . ".average_denominator_field");
count++;
);
);
);
# debug_message("export query header after sum setup: \n" . node_as_string(query_header) . "\n\n");
if (handle_aggregation) then (
# ------------------------------------
# Set aggreagtion values in query_data
# ------------------------------------
debug_message("\n\n" . node_as_string("volatile.aggregation_columns_header") . "\n\n");
node item;
node column;
string field_name;
string denominator_field;
float field_value;
float denominator_value;
float average_value;
foreach item query_data (
foreach column "volatile.aggregation_columns_header" (
field_name = node_value(subnode_by_name(column, "field_name"));
denominator_field = node_value(subnode_by_name(column, "average_denominator_field"));
field_value = node_value(subnode_by_name(item, field_name));
denominator_value = node_value(subnode_by_name(item, denominator_field));
average_value = field_value / denominator_value;
# debug_message("calculate aggregation --> $field_value / $denominator_value = $average_value \n");
set_node_value(subnode_by_name(item, field_name), average_value);
);
);
);
# -----------------------------------------------
# Sort the query as in the report element display
# -----------------------------------------------
string sort_field = "";
string sort_direction = "";
string sort_type;
string sort_method;
if (node_exists(report_element_node . ".sort_by") and (node_value(report_element_node . ".sort_by") ne "")) then (
sort_field = node_value(report_element_node . ".sort_by");
sort_direction = (if node_exists(report_element_node . ".sort_direction") then $(report_element_node . ".sort_direction") else "descending");
sort_type = get_sort_type(profile_name, sort_field);
sort_method = "field:" . sort_field . "," . sort_type . "," . sort_direction;
sort(query_data, sort_method);
);
# --------------
# number of rows
# --------------
# Note, due the huge amount a log_detail report doesn't show all records in the query_data node
# The number of rows in log data depends of the current set ending_row, if set to 50, the 50 rows
# will be in the query data.
int num_total_rows = num_subnodes(query_data);
int num_export_rows = num_total_rows;
if (node_exists("internal.command_line.ending_row") and
(node_value("internal.command_line.ending_row") ne "")) then (
int i = node_value("internal.command_line.ending_row");
if (i <= num_total_rows) then (
num_export_rows = i;
);
);
# debug_message("EXPORT number of rows query_data: " . num_total_rows);
# -------------
# csv_delimiter
# -------------
string csv_delimiter = ",";
if (node_exists(profile_path . ".statistics.miscellaneous.csv_delimiter")) then (
string dat = node_value(profile_path . ".statistics.miscellaneous.csv_delimiter");
if (dat ne "") then (
csv_delimiter = dat;
);
);
int csv_delimiter_int = char_to_ascii(csv_delimiter);
string csv_delimiter_in_hex = format(csv_delimiter_int, '%X');
csv_delimiter_in_hex = "%" . csv_delimiter_in_hex;
debug_message("#### CSV delimiter: " . csv_delimiter . "\n");
debug_message("#### CSV delimiter in hex: " . csv_delimiter_in_hex . "\n");
# ------------------
# end_of_line_marker
# ------------------
string end_of_line_marker;
if (node_exists("internal.export.end_of_line") and
(node_value("internal.export.end_of_line") ne "")) then (
end_of_line_marker = node_value("internal.export.end_of_line");
)
else (
if (_PLATFORM eq "Win32") then (
end_of_line_marker = "\r\n";
)
else (
end_of_line_marker = "\n";
);
);
# ---
# CSV
# ---
node row;
node column;
string main_field_node_name;
string column_label;
string column_type;
string column_field_name;
string column_field_type;
string column_display_format_type;
string column_value;
node column_sum_node;
int last_column_number = num_subnodes(query_header) - 1;
int column_number;
bool contains_delimiter;
bool contains_double_quote;
string csv_contents = "";
# ----------
# CSV header
# ----------
column_number = 0;
foreach column query_header (
column_label = node_value(subnode_by_name(column, "header_label"));
column_label = capitalize(expand(column_label));
if (column_number != last_column_number) then (
csv_contents .= column_label . csv_delimiter;
)
else (
csv_contents .= column_label . end_of_line_marker;
);
column_number++;
); # foreach column query_header
# --------
# CSV data
# --------
# If we don't need to show/calculate the totals row set num_total_rows to num_export_rows
if (!export_total_row and (num_export_rows != num_total_rows)) then (
num_total_rows = num_export_rows;
);
for (int row_number = 1; row_number <= num_total_rows; row_number++) (
# Get the query row node for this table row
row = subnode_by_number(query_data, row_number - 1);
column_number = 0;
foreach column query_header (
column_type = node_value(subnode_by_name(column, "type"));
column_field_name = node_value(subnode_by_name(column, "field_name"));
column_field_type = node_value(subnode_by_name(column, "data_type"));
column_value = node_value(subnode_by_name(row, column_field_name));
if (row_number <= num_export_rows) then (
# add data to csv
# ------------------------
# Decode __HexEsc__ values
# ------------------------
if (contains(column_value, "__HexEsc__")) then (
column_value = replace_all(column_value, "__HexEsc__", "%");
);
# ----------------------------------
# Escape delimiter and double quotes
# ----------------------------------
# debug_message("#### #### export value before escape: " . column_value . "\n");
contains_delimiter = contains(column_value, csv_delimiter_in_hex);
contains_double_quote = contains(column_value, "%22");
if (contains_delimiter or contains_double_quote) then (
# enclose column_value with double quotes!
column_value = '"' . column_value . '"';
# debug_message("#### #### escaped export value: " . column_value . "\n");
);
# ----------------
# Write csv string
# ----------------
csv_contents .= column_value;
if (column_number != last_column_number) then (
csv_contents .= csv_delimiter;
)
else (
csv_contents .= end_of_line_marker;
);
); # row_number <= num_export_rows
if (export_total_row) then (
# -----------------
# add values to sum
# -----------------
if (subnode_exists(column, "sum")) then (
column_sum_node = subnode_by_name(column, "sum");
set_node_value(column_sum_node, node_value(column_sum_node) + column_value);
);
);
column_number++;
); # foreach column query_header
); # for rows
# debug_message("export query header after sum calculation: \n" . node_as_string(query_header) . "\n\n");
# -------------
# Add total row
# -------------
if (export_total_row) then (
int column_number = 0;
string total_row = lang_stats.table.total_label . csv_delimiter;
foreach column query_header (
if (column_number > 0) then (
if (subnode_exists(column, "sum")) then (
total_row .= node_value(subnode_by_name(column, "sum")) . csv_delimiter ;
)
else (
total_row .= csv_delimiter;
);
);
column_number++;
);
# add end of line
total_row = replace_last(total_row, csv_delimiter, end_of_line_marker);
csv_contents .= total_row;
);
csv_contents = convert_escapes(csv_contents);
# If charset conversion is enabled, do the conversion now
if (node_exists("profiles." . profile_name . ".output.convert_export_charset") and node_value("profiles." . profile_name . ".output.convert_export_charset")) then (
debug_message("Converted CSV export from " . output.convert_export_from_charset . " to " . output.convert_export_to_charset . "\n");
csv_contents = convert_charset(csv_contents, output.convert_export_from_charset, output.convert_export_to_charset);
debug_message("Done converting CSV export\n");
);
if (export_to_file) then ( # via Web GUI
# ----------------------
# Get CSV directory path
# ----------------------
string divider = "/";
string user_node_name = "none";
bool webserver_mode = check_webserver_mode();
if (node_exists("volatile.authenticated_user_node_path")) then (
user_node_name = node_name(volatile.authenticated_user_node_path);
);
string csv_directory_path;
string csv_sub_directory_path = "csv_export" . divider . user_node_name . divider . profile_name;
if (webserver_mode) then (
string log_analysis_info_directory = get_loganalysisinfo_dir();
debug_message("#### log_analysis_info_directory: " . log_analysis_info_directory . "\n");
csv_directory_path = log_analysis_info_directory . "WebServerRoot" . divider . csv_sub_directory_path;
)
else (
# cgi-mode
csv_directory_path = node_value("preferences.server.temporary_directory_pathname");
if (!ends_with(csv_directory_path, divider)) then (
csv_directory_path .= divider;
);
csv_directory_path .= csv_sub_directory_path;
);
debug_message("#### csv_directory_path: " . csv_directory_path . "\n");
# -----------------
# Get csv_file_name
# -----------------
# Make file names maximal 32 characters
string report_part_of_file_name;
string csv_filename;
# int date_time_stamp_in_sec = now();
# debug_message("#### date_time_stamp_in_sec: " . date_time_stamp_in_sec . "\n");
if (report_name eq report_element) then (
# csv_filename = report_element . "_" . date_time_stamp_in_sec . ".csv";
report_part_of_file_name = report_element;
)
else (
# csv_filename = report_name . "_" . report_element . "_" . date_time_stamp_in_sec . ".csv";
report_part_of_file_name = report_name . "_" . report_element;
);
if (length(report_part_of_file_name) > 24) then (
report_part_of_file_name = substr(report_part_of_file_name, 0, 24);
);
string csv_directory_path_with_filename;
int count = 1;
string file_number;
bool is_unique_file_name = false;
while (!is_unique_file_name) (
if (count > 99) then (
file_number = count;
)
else if (count > 9) then (
file_number = "0" . count;
)
else (
file_number = "00" . count;
);
csv_filename = report_part_of_file_name . "_" . file_number . ".csv";
csv_directory_path_with_filename = csv_directory_path . "/" . csv_filename;
if (!file_exists(csv_directory_path_with_filename)) then (
is_unique_file_name = true;
);
count++;
);
# debug_message("#### csv_filename: " . csv_filename . "\n");
# ----------
# Write file
# ----------
write_file(csv_directory_path_with_filename, csv_contents);
# -----------------
# Set download link
# -----------------
string csv_export_path = "csv_export/" . user_node_name . "/" . profile_name . "/" . csv_filename;
string download_href;
if (webserver_mode) then (
# Note, this includes cgi-mode in case that
# preferences.server.temporary_directory_url and
# preferences.server.temporary_directory_pathname is empty!
# This is the new cgi-mode!
download_href = 'href=' . fileref(csv_export_path);
)
else (
# cgi-mode (old, see above for new cgi-mode)
string temporary_directory_url = node_value("preferences.server.temporary_directory_url");
if (!ends_with(temporary_directory_url, "/")) then (
temporary_directory_url .= "/";
);
download_href = 'href="' . temporary_directory_url . csv_export_path . '"';
);
debug_message("#### download_href: " . download_href . "\n");
'
\n';
'