For example, in the sentence:
That cake is extremely nice.
The rules of the English language would make cake a noun, extremely an adverb that modifies the adjective nice, and through this analysis the meaning could be understood.
However, this analysis is dependent on us recognising that the sequence of symbols used are words. If the characters used were not familiar to us we would not be able to do this. If we encountered a sentence using an unfamiliar notation, such as Chinese, parsing in this manner might be difficult. Here is an example Chinese sentence:
我可以写一点儿中文。
For anyone who does not read Chinese characters, it would not be clear which symbols combined to form words. The same could be true for a computer algorithm when processing either English or Chinese.
Thus parsing must be proceeded by a process known as lexical analysis or scanning, where the individual characters are grouped together into recognised symbols, which we might commonly call words, but in parsing algorithms are called tokens.