use super::{AnalyticsDirectory, AnalyticsMode, BatchProgramAnalyzerPlugin, BatchProgramAnalyzerContext}; use crate::common::create_csv_file; use crate::common::RecordBigram; use crate::common::RecordTrigram; use crate::common::RecordSkipgram; use crate::common::RecordUnigram; use loda_rust_core; use loda_rust_core::parser::ParameterType; use loda_rust_core::parser::{InstructionId, ParsedProgram}; use std::path::PathBuf; use std::error::Error; use std::collections::HashMap; type HistogramBigramKey = (String,String); type HistogramTrigramKey = (String,String,String); type HistogramSkipgramKey = (String,String); /// Creates [N-gram] csv files with lines of LODA code. /// /// Traverses all the programs inside the `loda-programs/oeis` dir. /// It looks for all the LODA assembly programs there are. /// Determines the most frequent combinations of LODA code lines. /// /// --- /// /// This outputs a `histogram_line_unigram.csv` file, with this format: /// /// ```csv /// count;word /// 96597;STOP /// 96597;START /// 69080;lpe /// 43970;lpb $0 /// 38553;mov $0,$1 /// 37459;add $0,1 /// 35716;sub $0,1 /// 21085;mov $1,$0 /// ``` /// /// Learnings from this unigram with LODA programs: /// /// Learning A: Loop instructions are the most used instructions. /// /// Learning B: As of 2022-oct-16 there are 102k LODA programs. /// When ignoring values greater than 100, /// then there are 9k rows that 2 or more times and 12k rows that occurs once. /// /// --- /// /// This outputs a `histogram_line_bigram.csv` file, with this format: /// /// ```csv /// count;word0;word1 /// 25134;mov $0,$1;STOP /// 20623;lpb $0;sub $0,1 /// 19807;lpe;mov $0,$1 /// 14278;START;mov $1,$0 /// 11410;add $0,1;STOP /// 10803;START;add $0,1 /// 10034;START;mov $1,1 /// 8582;START;lpb $0 /// ``` /// /// Learnings from this bigram with LODA programs: /// /// Learning A: Many programs ends with a `mov $0,$1` instruction. /// Many programs starts with a `mov $1,$0` instruction, which is the opposite. /// /// --- /// /// This outputs a `histogram_line_trigram.csv` file, with this format: /// /// ```csv /// count;word0;word1;word2 /// 12576;lpe;mov $0,$1;STOP /// 4255;mov $4,$0;max $4,0;cmp $4,$0 /// 4022;lpe;mov $0,$2;STOP /// 3823;add $0,1;lpb $0;sub $0,1 /// 3607;max $4,0;cmp $4,$0;mul $2,$4 /// 3464;lpe;mov $0,$3;STOP /// 3288;sub $2,1;lpe;mov $0,$1 /// 3197;mul $2,$4;sub $2,1;lpe /// ``` /// /// --- /// /// This outputs a `histogram_line_skipgram.csv` file, with this format: /// /// ```csv /// count;word0;word2 /// 27776;lpe;STOP /// 11992;START;lpb $0 /// 9280;mov $0,$1;STOP /// 5704;START;mov $2,$0 /// 5403;START;add $0,1 /// 5131;lpe;add $0,1 /// 4397;mov $4,$0;cmp $4,$0 /// 4249;add $0,1;sub $0,1 /// ``` /// /// [N-gram]: pub struct AnalyzeLineNgram { analytics_directory: AnalyticsDirectory, limit_lower: i64, limit_upper: i64, histogram_unigram: HashMap, histogram_bigram: HashMap, histogram_trigram: HashMap, histogram_skipgram: HashMap, ignore_count: usize, } impl AnalyzeLineNgram { pub fn new(analytics_directory: AnalyticsDirectory, mode: AnalyticsMode) -> Self { match mode { AnalyticsMode::OEIS => { // When mining for OEIS sequences, then magic constants are unwanted. // Values like 3, or 7 are considered nice values. // However a values like 123456789 is considered huge and magic. let limit_lower: i64 = -100; let limit_upper: i64 = 100; return Self::create(analytics_directory, limit_lower, limit_upper); }, AnalyticsMode::ARC => { // No particular preference about the lower limit. let limit_lower: i64 = -100; // It's a big memory layout of the ARC puzzles. // The input registers are between 100 and 200. // Thus the `magic_constant_upper_limit` is 200. let limit_upper: i64 = 200; return Self::create(analytics_directory, limit_lower, limit_upper); }, } } fn create(analytics_directory: AnalyticsDirectory, limit_lower: i64, limit_upper: i64) -> Self { Self { analytics_directory, limit_lower, limit_upper, histogram_unigram: HashMap::new(), histogram_bigram: HashMap::new(), histogram_trigram: HashMap::new(), histogram_skipgram: HashMap::new(), ignore_count: 0, } } fn extract_words(&mut self, parsed_program: &ParsedProgram) -> Vec { let mut words: Vec = vec!(); words.push("START".to_string()); for instruction in &parsed_program.instruction_vec { // Ignore programs that serves as bad examples. Huge magic constants are unwanted. // When there is a huge magic constant, then ignore the program. // For the `seq` instruction allow huge constants. // For the unofficial `fxy` instructions allow huge constants. let should_reject_extreme_source_constant: bool = match instruction.instruction_id { InstructionId::EvalSequence | InstructionId::UnofficialFunction { .. } => false, _ => true }; if should_reject_extreme_source_constant { if instruction.parameter_vec.len() == 2 { if let Some(parameter) = instruction.parameter_vec.last() { if parameter.parameter_type == ParameterType::Constant { if parameter.parameter_value < self.limit_lower { debug!("Encountered a magic value that is lower than {}. Ignoring program. Instruction: {}", self.limit_lower, instruction); self.ignore_count += 1; return vec!(); } if parameter.parameter_value > self.limit_upper { debug!("Encountered a magic value that is higher than {}. Ignoring program. Instruction: {}", self.limit_upper, instruction); self.ignore_count += 1; return vec!(); } } } } } // Data seems good, append it. let word: String = format!("{}", instruction); words.push(word); } words.push("STOP".to_string()); words } fn populate_unigram(&mut self, words: &Vec) { let keys: Vec = words.clone(); for key in keys { let counter = self.histogram_unigram.entry(key).or_insert(0); *counter += 1; } } fn populate_bigram(&mut self, words: &Vec) { let mut keys = Vec::::new(); let mut prev_word = String::new(); for (index, word1) in words.iter().enumerate() { let word0: String = prev_word; prev_word = word1.clone(); if index == 0 { continue; } let key: HistogramBigramKey = (word0, word1.clone()); keys.push(key); } for key in keys { let counter = self.histogram_bigram.entry(key).or_insert(0); *counter += 1; } } fn populate_trigram(&mut self, words: &Vec) { let mut keys = Vec::::new(); let mut prev_prev_word = String::new(); let mut prev_word = String::new(); for (index, word2) in words.iter().enumerate() { let word0: String = prev_prev_word; let word1: String = prev_word.clone(); prev_prev_word = prev_word; prev_word = word2.clone(); if index < 2 { continue; } let key: HistogramTrigramKey = (word0, word1, word2.clone()); keys.push(key); } for key in keys { let counter = self.histogram_trigram.entry(key).or_insert(0); *counter += 1; } } fn populate_skipgram(&mut self, words: &Vec) { let mut keys = Vec::::new(); let mut prev_prev_word = String::new(); let mut prev_word = String::new(); for (index, word2) in words.iter().enumerate() { let word0: String = prev_prev_word; prev_prev_word = prev_word; prev_word = word2.clone(); if index < 2 { continue; } let key: HistogramSkipgramKey = (word0, word2.clone()); keys.push(key); } for key in keys { let counter = self.histogram_skipgram.entry(key).or_insert(0); *counter += 1; } } fn save_unigram(&self) -> Result<(), Box> { // Convert from dictionary to array let mut records = Vec::::new(); for (histogram_key, histogram_count) in &self.histogram_unigram { let record = RecordUnigram { count: *histogram_count, word: histogram_key.clone(), }; records.push(record); } // Move the most frequently occuring items to the top // Move the lesser used items to the bottom records.sort_unstable_by_key(|item| (item.count, item.word.clone())); records.reverse(); // Save as a CSV file let output_path: PathBuf = self.analytics_directory.histogram_line_unigram_file(); create_csv_file(&records, &output_path) } fn save_bigram(&self) -> Result<(), Box> { // Convert from dictionary to array let mut records = Vec::::new(); for (histogram_key, histogram_count) in &self.histogram_bigram { let record = RecordBigram { count: *histogram_count, word0: histogram_key.0.clone(), word1: histogram_key.1.clone() }; records.push(record); } // Move the most frequently occuring items to the top // Move the lesser used items to the bottom records.sort_unstable_by_key(|item| (item.count, item.word0.clone(), item.word1.clone())); records.reverse(); // Save as a CSV file let output_path: PathBuf = self.analytics_directory.histogram_line_bigram_file(); create_csv_file(&records, &output_path) } fn save_trigram(&self) -> Result<(), Box> { // Convert from dictionary to array let mut records = Vec::::new(); for (histogram_key, histogram_count) in &self.histogram_trigram { let record = RecordTrigram { count: *histogram_count, word0: histogram_key.0.clone(), word1: histogram_key.1.clone(), word2: histogram_key.2.clone() }; records.push(record); } // Move the most frequently occuring items to the top // Move the lesser used items to the bottom records.sort_unstable_by_key(|item| (item.count, item.word0.clone(), item.word1.clone(), item.word2.clone())); records.reverse(); // Save as a CSV file let output_path: PathBuf = self.analytics_directory.histogram_line_trigram_file(); create_csv_file(&records, &output_path) } fn save_skipgram(&self) -> Result<(), Box> { // Convert from dictionary to array let mut records = Vec::::new(); for (histogram_key, histogram_count) in &self.histogram_skipgram { let record = RecordSkipgram { count: *histogram_count, word0: histogram_key.0.clone(), word2: histogram_key.1.clone() }; records.push(record); } // Move the most frequently occuring items to the top // Move the lesser used items to the bottom records.sort_unstable_by_key(|item| (item.count, item.word0.clone(), item.word2.clone())); records.reverse(); // Save as a CSV file let output_path: PathBuf = self.analytics_directory.histogram_line_skipgram_file(); create_csv_file(&records, &output_path) } } impl BatchProgramAnalyzerPlugin for AnalyzeLineNgram { fn plugin_name(&self) -> &'static str { "AnalyzeLineNgram" } fn analyze(&mut self, context: &BatchProgramAnalyzerContext) -> Result<(), Box> { let words: Vec = self.extract_words(&context.parsed_program); self.populate_unigram(&words); self.populate_bigram(&words); self.populate_trigram(&words); self.populate_skipgram(&words); Ok(()) } fn save(&self) -> Result<(), Box> { self.save_unigram()?; self.save_bigram()?; self.save_trigram()?; self.save_skipgram()?; Ok(()) } fn human_readable_summary(&self) -> String { let items: Vec = vec![ format!("unigram: {:?}", self.histogram_unigram.len()), format!("bigram: {:?}", self.histogram_bigram.len()), format!("trigram: {:?}", self.histogram_trigram.len()), format!("skipgram: {:?}", self.histogram_skipgram.len()), format!("ignore count: {:?}", self.ignore_count), ]; items.join(", ") } }