uva 907 pku 1093 formatting text

 

Formatting Text Time Limit: 1000MS   Memory Limit: 10000K Total Submissions: 1099   Accepted: 239 Description Writings e-mails is fun, but, unfortunately, they do not look very nice, mainly because not all lines have the same lengths. In this problem, your task is to write an e-mail formatting program which reformats a paragraph of an e-mail (e.g. by inserting spaces) so that, afterwards, all lines have the same length (even the last one of each paragraph).  The easiest way to perform this task would be to insert more spaces between the words in lines which are too short. But this is not the best way. Consider the following example:  **************************** This is the example you are actually considering. Let us assume that we want to get lines as long as the row of stars. Then, by simply inserting spaces, we would get  **************************** This is the example you are actually considering. But this looks rather odd because of the big gap in the second line. By moving the word ``are'' from the first to the second line, we get a better result:  **************************** This is the example you are actually considering. Of course, this has to be formalized. To do this, we assign a badness to each gap between words. The badness assigned to a gap of n spaces is (n - 1)^2. The goal of the program is to minimize the sum of all badnesses. For example, the badness of the first example is 1 + 7^2 = 50 whereas the badness of the second one is only 1 + 1 + 1 + 4 + 1 + 4 = 12.  In the output, every line has to start and to end with a word. (I.e. there cannot be a gap at the beginning or the end of a line.) The only exception to this is the following:  If a line contains only one word this word shall be put at the beginning of the line, and a badness of 500 is assigned to this line if it is shorter than it should be. (Of course, in this case, the length of the line is simply the length of the word.)  Input The input contains a text consisting of several paragraphs. Each paragraph is preceded by a line containing a single integer n, the desired width of the paragraph (1 <= n <= 80).  Paragraphs consist of one or more lines which contain one or more words each. Words consist of characters with ASCII codes between 33 and 126, inclusive, and are separated by spaces (possibly more than one). No word will be longer than the desired width of the paragraph. The total length of all words of one paragraph will not be more than 10000 characters.  Each paragraph is terminated by exactly one blank line. There is no limit on the number of paragraphs in the input file.  The input file will be terminated by a paragraph description starting with n=0. This paragraph should not be processed.  Output Output the same text, formatted in the way described above (processing each paragraph separately).  If there are several ways to format a paragraph with the same badness, use the following algorithm to choose which one to output: Let A and B be two solutions. Find the first gap which has not the same length in A and B. Do not output the solution in which this gap is bigger.  Output a blank line after each paragraph.  Sample Input 28 This is the example you are actually considering. 25 Writing e-mails is fun, and with this program, they even look nice. 0 Sample Output This is the example you are actually considering. Writing e-mails is fun, and with this program, they even look nice. Source Mid-Central European Regional Contest 1999

 

一直在自己的电脑上检查检查，在pku上是MLE，在uva上是RE，后来把数组开小一点，AC了，呵呵，真是好无语。

 

用动态规划解的：

 

mb[a, b] 表示 行a 覆盖到 单词b 的 最小Badness（Minmum Badness）

 

b[c, d] 表示 单词c 到 单词d 单独占一行的 Badness。

 

mb[ln, wd] = min{mb[ln - 1, i] + b[i + 1][wd] | 0<=i<wd}

 

边界值：

 

mb[ln, wd] = b[0, wd],  ln == 0

mb[ln, wd] = 500,         ln == wd

 

代码中 mb[][] 用 minBad[][] 数组保存，b[][] 用 bad[][]数组保存。

 

#include <cstdio> #include <cstdlib> #include <cassert> #include <cstring> #include <algorithm> #include <vector> #include <string> using namespace std; #define INT_INF 0x0fffffff #define WORD_MAX 900 #define WORD_LEN_MAX 81 #define LINE_MAX 900 char word[WORD_MAX][WORD_LEN_MAX]; int minBad[LINE_MAX][WORD_MAX]; // mb[] int bad[WORD_MAX][WORD_MAX]; // b[] int wdLenSum[WORD_MAX]; // wdLenSum[i] 保存 word[0] 到 word[i] 的单词长度和 int lineWd[LINE_MAX][WORD_MAX]; // 输出时用，lineWd[ln][wd] 保存策略为 ln，wd 时该行单词数 int wordSz; // 单词数 int lineLen; // 行长度 int lineSz; // 行数 int DP() { // 初始化 wdLenSum[] fill(wdLenSum, wdLenSum + wordSz, 0); for(int w=0; w<wordSz; ++w) { int wdLen = strlen(word[w]); for(int we=w; we<wordSz; ++we) wdLenSum[we] += wdLen; } // 初始化 bad[][] for(int w=0; w<wordSz; ++w) { if((w > 0) && (wdLenSum[w] - wdLenSum[w - 1] == lineLen) || (w == 0) && (wdLenSum[w] == lineLen)) bad[w][w] = 0; else bad[w][w] = 500; } for(int w=0; w<wordSz; ++w) { for(int we=w+1; we<wordSz; ++we) { int a = lineLen - wdLenSum[we]; if(w > 0) a += wdLenSum[w - 1]; int n = we - w; int b = -2 * a + n; int r = a % n; int t = a / n; if(a < n) bad[w][we] = INT_INF; else bad[w][we] = b + t * t * (n - r) + (t + 1) * (t + 1) * r; } } // 初始化 minBad 边界值 for(int ln=0; ln<lineSz; ++ln) { for(int wd=0; wd<wordSz; ++wd) { if(ln == 0) { minBad[ln][wd] = bad[0][wd]; lineWd[ln][wd] = wd + 1; } else if(ln <= wd) { minBad[ln][wd] = minBad[ln - 1][ln - 1] + bad[ln][wd]; lineWd[ln][wd] = 1; } else { minBad[ln][wd] = INT_INF; lineWd[ln][wd] = 0; } } } // 开始递推 for(int ln=1; ln<lineSz; ++ln) { for(int wd=ln; wd<wordSz; ++wd) { for(int i=ln-1; i<wd; ++i) { int newBad = minBad[ln - 1][i] + bad[i + 1][wd]; if(newBad <= minBad[ln][wd]) { minBad[ln][wd] = newBad; // mark trace lineWd[ln][wd] = wd - i; } } } } #ifdef _DEBUG printf("minBad:/n"); for(int ln=0; ln<lineSz; ++ln) { for(int wd=0; wd<wordSz; ++wd) printf(wd==wordSz-1 ? "%5d/n" : "%5d", wd>=ln && minBad[ln][wd]<INT_INF ? minBad[ln][wd] : -1); } #endif return 0; } vector <string> toPrint; void PrintFmttedTxt() { // 找出最优策略的 行line int line = lineSz - 1; for(int ln=lineSz-1; ln>=0; --ln) { if(minBad[line][wordSz - 1] >= minBad[ln][wordSz - 1]) line = ln; } int wd = wordSz; // 剩余单词数 toPrint.clear(); while(line >= 0 && wd > 0) // 每次循环打印一行 存入 toPrint 的最后 { int wdCnt = lineWd[line][wd - 1]; // 该行单词数 static char buf[20001] = ""; strcpy(buf, ""); // 输出每个单词和随后的空格，除了最后一个单词 for(int w=0; w<wdCnt-1; ++w) { sprintf(buf + strlen(buf), "%s", word[w + wd - wdCnt]); // 计算空格数 int spCnt = lineLen - wdLenSum[wd - 1]; if(wd - wdCnt > 0) spCnt += wdLenSum[wd - wdCnt - 1]; spCnt = (spCnt + w) / (wdCnt - 1); // 打印空格 for(int sp=0; sp<spCnt; ++sp) sprintf(buf + strlen(buf), " "); } // 每行最后一个单词 if(wd > 0) sprintf(buf + strlen(buf), "%s", word[wd - 1]); toPrint.push_back(buf); wd -= wdCnt; --line; } // 逆序打印出来 for(int i=toPrint.size()-1; i>=0; --i) printf("%s/n", toPrint[i].c_str()); } int main() { while(true) { scanf("%d", &lineLen); if(lineLen == 0) return 0; wordSz = 0; static char buf[20001]; gets(buf); while(gets(buf) != NULL && strlen(buf) > 0) { char *p = strtok(buf, " /t/n"); while(p != NULL) { sscanf(p, "%s", word[wordSz]); if(strlen(word[wordSz]) > 0) ++wordSz; p = strtok(NULL, " /t/n"); } } lineSz = wordSz; DP(); PrintFmttedTxt(); printf("/n"); } return 0; }